udacity-flowers-pytorch-daya.ipynb

{"cells":[{"cell_type":"markdown","metadata":{},"source":["## Prepare the workspace"]},{"cell_type":"code","execution_count":1,"metadata":{"scrolled":true},"outputs":[{"name":"stdout","output_type":"stream","text":["2.1.2\n","False\n","cpu\n","Using device: mps\n"]}],"source":["# Check torch version and CUDA status if GPU is enabled.\n","import torch\n","print(torch.__version__)\n","print(torch.cuda.is_available()) # Should return True when GPU is enabled. \n","device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')\n","print(device)\n","\n","# Set the device      \n","device = \"mps\" if torch.backends.mps.is_available() else \"cpu\"\n","print(f\"Using device: {device}\")"]},{"cell_type":"markdown","metadata":{},"source":["# Developing an AI application\n","\n","Going forward, AI algorithms will be incorporated into more and more everyday applications. For example, you might want to include an image classifier in a smart phone app. To do this, you'd use a deep learning model trained on hundreds of thousands of images as part of the overall application architecture. A large part of software development in the future will be using these types of models as common parts of applications. \n","\n","In this project, you'll train an image classifier to recognize different species of flowers. You can imagine using something like this in a phone app that tells you the name of the flower your camera is looking at. In practice you'd train this classifier, then export it for use in your application. We'll be using [this dataset](http://www.robots.ox.ac.uk/~vgg/data/flowers/102/index.html) of 102 flower categories, you can see a few examples below. \n","\n","<img src='assets/Flowers.png' width=500px>\n","\n","The project is broken down into multiple steps:\n","\n","* Load and preprocess the image dataset\n","* Train the image classifier on your dataset\n","* Use the trained classifier to predict image content\n","\n","We'll lead you through each part which you'll implement in Python.\n","\n","When you've completed this project, you'll have an application that can be trained on any set of labeled images. Here your network will be learning about flowers and end up as a command line application. But, what you do with your new skills depends on your imagination and effort in building a dataset. For example, imagine an app where you take a picture of a car, it tells you what the make and model is, then looks up information about it. Go build your own dataset and make something new.\n","\n","First up is importing the packages you'll need. It's good practice to keep all the imports at the beginning of your code. As you work through this notebook and find you need to import a package, make sure to add the import up here."]},{"cell_type":"code","execution_count":2,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:50:27.956808Z","iopub.status.busy":"2023-12-30T11:50:27.956415Z","iopub.status.idle":"2023-12-30T11:50:30.827550Z","shell.execute_reply":"2023-12-30T11:50:30.826658Z","shell.execute_reply.started":"2023-12-30T11:50:27.956772Z"},"trusted":true},"outputs":[],"source":["# Imports here\n","\n","import torch\n","from torchvision import transforms, datasets, models\n","from torch import nn\n","import torch.optim.lr_scheduler as lr_scheduler\n","#from torchsummary import summary\n","\n","import json\n","from tqdm import tqdm\n","\n","from PIL import Image\n","import matplotlib.pyplot as plt\n","import numpy as np"]},{"cell_type":"markdown","metadata":{},"source":["## Load the data\n","\n","Here you'll use `torchvision` to load the data ([documentation](http://pytorch.org/docs/0.3.0/torchvision/index.html)). The data should be included alongside this notebook, otherwise you can [download it here](https://s3.amazonaws.com/content.udacity-data.com/nd089/flower_data.tar.gz). "]},{"cell_type":"markdown","metadata":{},"source":["If you do not find the `flowers/` dataset in the current directory, **/workspace/home/aipnd-project/**, you can download it using the following commands. \n","\n","```bash\n","!wget 'https://s3.amazonaws.com/content.udacity-data.com/nd089/flower_data.tar.gz'\n","!unlink flowers\n","!mkdir flowers && tar -xzf flower_data.tar.gz -C flowers\n","```\n"]},{"cell_type":"markdown","metadata":{},"source":["## Data Description\n","The dataset is split into three parts, training, validation, and testing. For the training, you'll want to apply transformations such as random scaling, cropping, and flipping. This will help the network generalize leading to better performance. You'll also need to make sure the input data is resized to 224x224 pixels as required by the pre-trained networks.\n","\n","The validation and testing sets are used to measure the model's performance on data it hasn't seen yet. For this you don't want any scaling or rotation transformations, but you'll need to resize then crop the images to the appropriate size.\n","\n","The pre-trained networks you'll use were trained on the ImageNet dataset where each color channel was normalized separately. For all three sets you'll need to normalize the means and standard deviations of the images to what the network expects. For the means, it's `[0.485, 0.456, 0.406]` and for the standard deviations `[0.229, 0.224, 0.225]`, calculated from the ImageNet images.  These values will shift each color channel to be centered at 0 and range from -1 to 1.\n"," "]},{"cell_type":"code","execution_count":3,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:15:40.737713Z","iopub.status.busy":"2023-12-30T11:15:40.737242Z","iopub.status.idle":"2023-12-30T11:15:40.743917Z","shell.execute_reply":"2023-12-30T11:15:40.742606Z","shell.execute_reply.started":"2023-12-30T11:15:40.737670Z"},"trusted":true},"outputs":[],"source":["data_dir = '/Users/dalex/Udacity/new_try/aipnd-project/flower_data'\n","train_dir = data_dir + '/train'\n","valid_dir = data_dir + '/valid'\n","test_dir = data_dir + '/test'"]},{"cell_type":"code","execution_count":4,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:15:43.843753Z","iopub.status.busy":"2023-12-30T11:15:43.843174Z","iopub.status.idle":"2023-12-30T11:15:48.688141Z","shell.execute_reply":"2023-12-30T11:15:48.687065Z","shell.execute_reply.started":"2023-12-30T11:15:43.843701Z"},"trusted":true},"outputs":[],"source":["# TODO: Define your transforms for the training, validation, and testing sets\n","train_transforms = transforms.Compose([\n","                                      transforms.RandomResizedCrop(size= (224,224)),\n","                                      transforms.RandomHorizontalFlip(p= 0.5),\n","                                      transforms.ToTensor(),\n","                                      transforms.Normalize([0.485, 0.456, 0.406],\n","                                                          [0.229, 0.224, 0.225])\n","])\n","\n","test_transforms = transforms.Compose([transforms.RandomResizedCrop(size= (224,224)),\n","                                     transforms.ToTensor(),\n","                                     transforms.Normalize([0.485, 0.456, 0.406],\n","                                                          [0.229, 0.224, 0.225])    \n","])\n","\n","img_datasets = {}\n","img_datasets['train'] = datasets.ImageFolder(train_dir, transform = train_transforms)\n","img_datasets['valid'] = datasets.ImageFolder(valid_dir, transform = test_transforms)\n","img_datasets['test'] = datasets.ImageFolder(test_dir, transform = test_transforms)\n","\n","# TODO: Using the image datasets and the trainforms, define the dataloaders\n","trainloaders = torch.utils.data.DataLoader(img_datasets['train'], batch_size = 32, shuffle = True)\n","validloaders = torch.utils.data.DataLoader(img_datasets['valid'], batch_size = 16, shuffle = True)\n","testloaders = torch.utils.data.DataLoader(img_datasets['test'], batch_size = 16, shuffle = True)"]},{"cell_type":"code","execution_count":5,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:15:48.690323Z","iopub.status.busy":"2023-12-30T11:15:48.689976Z","iopub.status.idle":"2023-12-30T11:15:48.697166Z","shell.execute_reply":"2023-12-30T11:15:48.696019Z","shell.execute_reply.started":"2023-12-30T11:15:48.690289Z"},"trusted":true},"outputs":[{"name":"stdout","output_type":"stream","text":["Train 6552 images || number of batches 205 \n","Valid 818 images || number of batches 52 \n","Test 819 images || number of batches 52 \n"]}],"source":["print(f'Train {len(trainloaders.dataset)} images || number of batches {len(trainloaders)} ')\n","print(f'Valid {len(validloaders.dataset)} images || number of batches {len(validloaders)} ')\n","print(f'Test {len(testloaders.dataset)} images || number of batches {len(testloaders)} ')"]},{"cell_type":"markdown","metadata":{},"source":["### Label mapping\n","\n","You'll also need to load in a mapping from category label to category name. You can find this in the file `cat_to_name.json`. It's a JSON object which you can read in with the [`json` module](https://docs.python.org/2/library/json.html). This will give you a dictionary mapping the integer encoded categories to the actual names of the flowers."]},{"cell_type":"code","execution_count":6,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:15:51.785073Z","iopub.status.busy":"2023-12-30T11:15:51.784050Z","iopub.status.idle":"2023-12-30T11:15:51.803454Z","shell.execute_reply":"2023-12-30T11:15:51.802093Z","shell.execute_reply.started":"2023-12-30T11:15:51.785022Z"},"trusted":true},"outputs":[],"source":["\n","with open('/Users/dalex/Udacity/aipnd-project/cat_to_name.json', 'r') as f:\n","    cat_to_name = json.load(f)\n","    \n","#print('cat_to_name',cat_to_name)\n","\n","\n","class_to_idx = img_datasets['train'].class_to_idx\n","\n","#print('class_to_idx', class_to_idx)\n","idx_to_class={}\n","for  name, idx in class_to_idx.items():\n","    label = cat_to_name.get(name)  \n","    idx_to_class[idx] = label  \n","\n","#print('idx_to_class',idx_to_class)"]},{"cell_type":"markdown","metadata":{},"source":["# Building and training the classifier\n","\n","Now that the data is ready, it's time to build and train the classifier. As usual, you should use one of the pretrained models from `torchvision.models` to get the image features. Build and train a new feed-forward classifier using those features.\n","\n","We're going to leave this part up to you. Refer to [the rubric](https://review.udacity.com/#!/rubrics/1663/view) for guidance on successfully completing this section. Things you'll need to do:\n","\n","* Load a [pre-trained network](http://pytorch.org/docs/master/torchvision/models.html) (If you need a starting point, the VGG networks work great and are straightforward to use)\n","* Define a new, untrained feed-forward network as a classifier, using ReLU activations and dropout\n","* Train the classifier layers using backpropagation using the pre-trained network to get the features\n","* Track the loss and accuracy on the validation set to determine the best hyperparameters\n","\n","We've left a cell open for you below, but use as many as you need. Our advice is to break the problem up into smaller parts you can run separately. Check that each part is doing what you expect, then move on to the next. You'll likely find that as you work through each part, you'll need to go back and modify your previous code. This is totally normal!\n","\n","When training make sure you're updating only the weights of the feed-forward network. You should be able to get the validation accuracy above 70% if you build everything right. Make sure to try different hyperparameters (learning rate, units in the classifier, epochs, etc) to find the best model. Save those hyperparameters to use as default values in the next part of the project.\n","\n","One last important tip if you're using the workspace to run your code: To avoid having your workspace disconnect during the long-running tasks in this notebook, please read in the earlier page in this lesson called Intro to\n","GPU Workspaces about Keeping Your Session Active. You'll want to include code from the workspace_utils.py module.\n","\n","## Note for Workspace users: \n","If your network is over 1 GB when saved as a checkpoint, there might be issues with saving backups in your workspace. Typically this happens with wide dense layers after the convolutional layers. If your saved checkpoint is larger than 1 GB (you can open a terminal and check with `ls -lh`), you should reduce the size of your hidden layers and train again."]},{"cell_type":"code","execution_count":7,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:51:06.416617Z","iopub.status.busy":"2023-12-30T11:51:06.416219Z","iopub.status.idle":"2023-12-30T11:51:06.431217Z","shell.execute_reply":"2023-12-30T11:51:06.430074Z","shell.execute_reply.started":"2023-12-30T11:51:06.416580Z"},"trusted":true},"outputs":[],"source":["# TODO: Build and train your network\n","\n","def build_model(model_name='resnet50', hidden_layer = 1000):\n","    if model_name == 'resnet50':\n","    \n","        model = models.resnet50(pretrained=True)\n","\n","        for param in model.parameters():\n","            param.requires_grad = False\n","    \n","        in_features = model.fc.in_features\n","\n","        classifier = nn.Sequential(\n","                                  nn.BatchNorm1d(num_features=in_features),\n","                                  nn.Linear(in_features, hidden_layer),\n","                                  nn.ReLU(), \n","                                  nn.BatchNorm1d(num_features=hidden_layer), \n","                                  nn.Linear(hidden_layer,102),\n","                                  nn.LogSoftmax(dim = 1)  \n","                                )  \n","                                    \n","        model.fc = classifier\n","        \n","    elif model_name =='densenet_121':\n","        \n","        model = models.densenet121(pretrained=True)\n","        \n","        for param in model.parameters():\n","            param.requires_grad = False\n","\n","        in_features = model.classifier.in_features\n","\n","        classifier= nn.Sequential(\n","                                  nn.BatchNorm1d(num_features=in_features),\n","                                  nn.Linear(in_features, hidden_layer),\n","                                  nn.ReLU(),\n","                                  nn.BatchNorm1d(num_features=hidden_layer),\n","                                  nn.Dropout(0.2),  \n","                                  nn.Linear(hidden_layer,102),\n","                                  nn.LogSoftmax(dim = 1)  \n","                                ) \n","        \n","    \n","        model.classifier = classifier\n","    \n","    return model\n"]},{"cell_type":"code","execution_count":8,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:51:16.012250Z","iopub.status.busy":"2023-12-30T11:51:16.011313Z","iopub.status.idle":"2023-12-30T11:51:16.018021Z","shell.execute_reply":"2023-12-30T11:51:16.016791Z","shell.execute_reply.started":"2023-12-30T11:51:16.012197Z"},"scrolled":true,"trusted":true},"outputs":[],"source":["#specs, hyperparameters\n","model_name = 'densenet_121'\n","if model_name == 'densenet_121':\n","    hidden_layer = 1000\n","else:\n","    hidden_layer = 512\n","\n","epochs = 5\n","train_losses, valid_losses, test_losses, val_acc, test_acc = [],[],[],[],[]\n","criterion = nn.NLLLoss()\n","lr = 0.001\n"]},{"cell_type":"code","execution_count":9,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:51:17.147615Z","iopub.status.busy":"2023-12-30T11:51:17.147275Z","iopub.status.idle":"2023-12-30T11:51:19.589070Z","shell.execute_reply":"2023-12-30T11:51:19.587946Z","shell.execute_reply.started":"2023-12-30T11:51:17.147583Z"},"trusted":true},"outputs":[{"name":"stdout","output_type":"stream","text":["{'state': {}, 'param_groups': [{'lr': 0.001, 'betas': (0.9, 0.999), 'eps': 1e-08, 'weight_decay': 0, 'amsgrad': False, 'maximize': False, 'foreach': None, 'capturable': False, 'differentiable': False, 'fused': None, 'params': [0, 1, 2, 3, 4, 5, 6, 7]}]}\n"]},{"name":"stderr","output_type":"stream","text":["/Users/dalex/miniforge3/envs/mps/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n","  warnings.warn(\n","/Users/dalex/miniforge3/envs/mps/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=DenseNet121_Weights.IMAGENET1K_V1`. You can also use `weights=DenseNet121_Weights.DEFAULT` to get the most up-to-date weights.\n","  warnings.warn(msg)\n"]}],"source":["\n","new_model = build_model(model_name,hidden_layer)\n","\n","if model_name == 'densenet_121':\n","    optimizer = torch.optim.Adam(new_model.classifier.parameters(), lr = lr)\n","else:\n","    optimizer = torch.optim.Adam(new_model.fc.parameters(), lr = lr)\n","    \n","scheduler = lr_scheduler.ReduceLROnPlateau(optimizer, mode='min', factor=0.1, patience=3, verbose=True)\n","\n","print(optimizer.state_dict())\n"]},{"cell_type":"code","execution_count":10,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:16:34.685768Z","iopub.status.busy":"2023-12-30T11:16:34.685328Z","iopub.status.idle":"2023-12-30T11:16:34.702877Z","shell.execute_reply":"2023-12-30T11:16:34.701579Z","shell.execute_reply.started":"2023-12-30T11:16:34.685724Z"},"trusted":true},"outputs":[],"source":["#TRAINING and VALIDATION\n","\n","def train_model(model,epochs,trainloaders, validloaders, criterion, optimizer, scheduler, gpu = 'gpu'):\n","    \n","    if gpu == 'gpu':\n","        device = 'mps' if torch.backends.mps.is_available() else 'cuda'\n","    else:\n","        device = 'cpu'\n","    print('device is :', device) \n","    model.to(device)\n","    \n","    for e in range(epochs):\n","        running_train_loss = 0\n","    \n","        model.train()\n","        for images, labels in tqdm(trainloaders, desc=f'Epoch {e+1}/{epochs} - Training'):\n","            images, labels = images.to(device), labels.to(device)\n","\n","            optimizer.zero_grad()\n","            log_ps = model(images)\n","            loss = criterion(log_ps, labels) #NLLLoss\n","            loss.backward()\n","            optimizer.step()\n","            running_train_loss += loss.item()\n","\n","        # Validation loop after training is complete\n","        running_val_accuracy_per_epoch = 0\n","        running_val_loss = 0\n","    \n","        model.eval()\n","        with torch.no_grad():\n","            for images, labels in tqdm(validloaders, desc=f'Epoch {e+1}/{epochs} - Validation'):\n","                images, labels = images.to(device), labels.to(device)\n","\n","                \n","                log_ps = model(images)\n","                valid_loss = criterion(log_ps, labels)\n","                running_val_loss += valid_loss.item()\n","                prob_per_batch = torch.exp(log_ps)\n","                \n","                top_prob_per_batch, top_class_per_batch = prob_per_batch.topk(1, dim=1)\n","                truth_val = top_class_per_batch == labels.view(*top_class_per_batch.shape)\n","                running_val_accuracy_per_epoch += torch.mean(truth_val.type(torch.FloatTensor))\n","\n","        epoch_val_accuracy = running_val_accuracy_per_epoch / len(validloaders)\n","        train_loss_per_epoch = running_train_loss / len(trainloaders)\n","        val_loss_per_epoch = running_val_loss / len(validloaders)\n","\n","        scheduler.step(val_loss_per_epoch)\n","        \n","        train_losses.append(train_loss_per_epoch)\n","        valid_losses.append(val_loss_per_epoch)\n","        val_acc.append(epoch_val_accuracy)\n","        \n","        print(f'epoch: {e+1}/{epochs}')\n","        print(f'train_loss_per_epoch: {train_losses[-1]}')\n","        print(f'validation_loss_per_epoch: {valid_losses[-1]}')\n","        print(f'epoch_validation_accuracy: {epoch_val_accuracy.item() * 100}%')\n","        print(f'current learning rate: {optimizer.param_groups[0][\"lr\"]}')\n","        \n","    return model, train_losses, valid_losses, val_acc\n"]},{"cell_type":"code","execution_count":11,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:20:14.424802Z","iopub.status.busy":"2023-12-30T11:20:14.424381Z","iopub.status.idle":"2023-12-30T11:27:35.132672Z","shell.execute_reply":"2023-12-30T11:27:35.130960Z","shell.execute_reply.started":"2023-12-30T11:20:14.424754Z"},"trusted":true},"outputs":[{"name":"stdout","output_type":"stream","text":["device is : mps\n"]},{"name":"stderr","output_type":"stream","text":["Epoch 1/5 - Training: 100%|██████████| 205/205 [01:43<00:00,  1.97it/s]\n","Epoch 1/5 - Validation: 100%|██████████| 52/52 [00:13<00:00,  3.77it/s]\n"]},{"name":"stdout","output_type":"stream","text":["epoch: 1/5\n","train_loss_per_epoch: 1.101649439916378\n","validation_loss_per_epoch: 0.5042558196358956\n","epoch_validation_accuracy: 87.25961446762085%\n","current learning rate: 0.001\n"]},{"name":"stderr","output_type":"stream","text":["Epoch 2/5 - Training: 100%|██████████| 205/205 [01:39<00:00,  2.05it/s]\n","Epoch 2/5 - Validation: 100%|██████████| 52/52 [00:11<00:00,  4.69it/s]\n"]},{"name":"stdout","output_type":"stream","text":["epoch: 2/5\n","train_loss_per_epoch: 0.4455524174178519\n","validation_loss_per_epoch: 0.42853469050561005\n","epoch_validation_accuracy: 89.30288553237915%\n","current learning rate: 0.001\n"]},{"name":"stderr","output_type":"stream","text":["Epoch 3/5 - Training: 100%|██████████| 205/205 [01:40<00:00,  2.03it/s]\n","Epoch 3/5 - Validation: 100%|██████████| 52/52 [00:11<00:00,  4.69it/s]\n"]},{"name":"stdout","output_type":"stream","text":["epoch: 3/5\n","train_loss_per_epoch: 0.3657348736938907\n","validation_loss_per_epoch: 0.40859680558339906\n","epoch_validation_accuracy: 89.66346383094788%\n","current learning rate: 0.001\n"]},{"name":"stderr","output_type":"stream","text":["Epoch 4/5 - Training: 100%|██████████| 205/205 [01:40<00:00,  2.03it/s]\n","Epoch 4/5 - Validation: 100%|██████████| 52/52 [00:11<00:00,  4.69it/s]\n"]},{"name":"stdout","output_type":"stream","text":["epoch: 4/5\n","train_loss_per_epoch: 0.3429804965853691\n","validation_loss_per_epoch: 0.4282037371482987\n","epoch_validation_accuracy: 90.625%\n","current learning rate: 0.001\n"]},{"name":"stderr","output_type":"stream","text":["Epoch 5/5 - Training: 100%|██████████| 205/205 [01:41<00:00,  2.03it/s]\n","Epoch 5/5 - Validation: 100%|██████████| 52/52 [00:11<00:00,  4.65it/s]"]},{"name":"stdout","output_type":"stream","text":["epoch: 5/5\n","train_loss_per_epoch: 0.31327520189125363\n","validation_loss_per_epoch: 0.44918425753712654\n","epoch_validation_accuracy: 89.18269276618958%\n","current learning rate: 0.001\n"]},{"name":"stderr","output_type":"stream","text":["\n"]}],"source":["trained_model, train_losses, valid_losses, val_acc = train_model(\n","    new_model,epochs,trainloaders, validloaders,criterion, optimizer, scheduler,'gpu') "]},{"cell_type":"code","execution_count":12,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["{'state': {0: {'step': tensor(1025.), 'exp_avg': tensor([-0.9588, -0.6099, -0.6378,  ..., -1.0894, -0.5917, -1.3343],\n","       device='mps:0'), 'exp_avg_sq': tensor([0.2049, 0.3178, 0.3425,  ..., 0.8952, 1.0310, 1.0329], device='mps:0')}, 1: {'step': tensor(1025.), 'exp_avg': tensor([2.3475e+00, 5.4543e-07, 1.8688e+00,  ..., 1.9402e-06, 1.2604e+00,\n","        3.7101e+00], device='mps:0'), 'exp_avg_sq': tensor([2.8942, 0.3389, 1.9888,  ..., 0.7389, 1.7460, 1.8896], device='mps:0')}, 2: {'step': tensor(1025.), 'exp_avg': tensor([[-4.9583e-05, -4.6547e-05, -5.6430e-05,  ...,  5.8040e-05,\n","          1.3040e-05,  2.1063e-04],\n","        [-1.9917e-09, -1.1026e-09,  9.4689e-10,  ..., -1.7639e-09,\n","          2.2169e-10, -3.1994e-09],\n","        [ 1.2602e-05, -4.4429e-04,  5.0897e-04,  ...,  2.3459e-05,\n","         -4.5508e-04,  1.3409e-03],\n","        ...,\n","        [-8.1133e-09, -4.2477e-09, -1.1772e-09,  ..., -7.8299e-09,\n","         -1.8016e-08, -4.0244e-09],\n","        [-1.7030e-08, -5.7349e-10,  4.2482e-08,  ..., -4.2540e-08,\n","         -1.3429e-07, -1.8390e-08],\n","        [ 2.5651e-05,  2.1956e-04,  9.1736e-05,  ...,  2.6784e-04,\n","         -6.2020e-05,  1.9316e-05]], device='mps:0'), 'exp_avg_sq': tensor([[3.5362e-08, 9.2166e-07, 4.4878e-07,  ..., 2.7821e-06, 2.9439e-06,\n","         2.7141e-06],\n","        [2.4557e-07, 2.4814e-06, 1.0833e-06,  ..., 5.6787e-06, 7.5181e-06,\n","         4.4008e-06],\n","        [7.5334e-08, 1.2917e-06, 7.3033e-07,  ..., 4.3066e-06, 4.3478e-06,\n","         3.5052e-06],\n","        ...,\n","        [3.7118e-07, 1.4954e-06, 9.1962e-07,  ..., 4.3929e-06, 7.1956e-06,\n","         4.1751e-06],\n","        [1.4512e-07, 1.6481e-06, 1.5999e-06,  ..., 5.5007e-06, 1.0046e-05,\n","         4.6673e-06],\n","        [2.6350e-08, 8.4421e-07, 3.8292e-07,  ..., 2.5330e-06, 2.9920e-06,\n","         1.9961e-06]], device='mps:0')}, 3: {'step': tensor(1025.), 'exp_avg': tensor([-1.2862e-05,  2.7546e-09, -2.7698e-05, -3.2044e-09,  1.2943e-22,\n","         4.3897e-07,  3.1997e-16, -7.0674e-09,  1.9969e-22,  3.1159e-06,\n","        -5.7681e-12, -8.2017e-06, -4.6190e-04,  1.1203e-04, -1.7923e-10,\n","        -4.6642e-05,  4.7681e-04,  5.7558e-06,  2.7639e-05,  1.3456e-04,\n","        -1.0855e-04,  6.9386e-06, -3.2877e-05,  2.1952e-05, -4.4950e-06,\n","         1.8600e-05, -1.4206e-19,  7.0174e-22,  2.2902e-22,  9.3915e-05,\n","         2.8012e-04,  1.7944e-06, -1.6698e-04, -4.1562e-11,  5.1514e-05,\n","         2.0370e-04,  6.6077e-13,  1.1009e-05,  1.6718e-15, -2.6503e-10,\n","         1.1676e-08, -4.5684e-05, -1.2659e-04, -3.4162e-05,  3.4904e-05,\n","        -5.1730e-05,  5.2858e-06, -2.8032e-07,  3.6703e-28,  5.0479e-03,\n","        -1.0854e-15,  4.5706e-05,  8.6823e-05, -8.0684e-07,  1.7231e-04,\n","         1.4373e-05,  9.2798e-06,  1.5155e-09,  6.5278e-06,  6.5043e-05,\n","        -1.1458e-04, -2.8089e-06, -8.4621e-22,  1.0447e-04,  6.8159e-10,\n","         3.0324e-07,  2.7273e-06,  2.4671e-13, -3.2036e-05,  4.5048e-21,\n","         3.0154e-04,  1.2515e-04,  1.4876e-15,  3.6609e-05, -3.0525e-08,\n","         2.5123e-04,  4.7235e-06,  1.5812e-05, -1.8401e-04,  1.2142e-05,\n","        -1.0553e-04,  2.1489e-05, -1.1778e-05, -3.1084e-10,  2.9747e-18,\n","        -1.1275e-06,  1.0039e-21,  6.4485e-27,  5.3256e-06,  9.4700e-05,\n","         2.1901e-04,  8.2297e-11, -1.3827e-05,  1.8170e-05,  1.7761e-04,\n","         1.0442e-05, -1.3853e-05,  1.4120e-05,  7.7244e-05,  5.2955e-14,\n","        -2.2228e-04, -2.1696e-05,  3.8039e-20,  1.6827e-07,  2.7518e-07,\n","         2.0249e-13, -1.7220e-04, -5.3752e-07,  7.4420e-05,  2.8712e-05,\n","        -1.4224e-06,  2.0725e-14, -3.4648e-05, -3.7265e-05,  2.0389e-10,\n","        -1.6243e-04,  1.1571e-05,  3.5151e-05,  3.6144e-21,  1.6779e-05,\n","         1.0609e-05,  6.5776e-07,  2.2412e-06,  5.3865e-13, -2.1345e-08,\n","         4.3244e-06, -8.5380e-06,  3.4148e-04,  2.5077e-05, -4.9503e-06,\n","         7.4560e-05, -3.3336e-05, -1.0315e-04, -8.6795e-05,  1.6284e-05,\n","        -2.3397e-05,  3.9767e-05,  1.9118e-16,  2.1196e-08, -2.9749e-05,\n","        -1.8769e-04, -2.2920e-04, -2.3383e-06,  4.3631e-05, -1.3709e-05,\n","        -1.1495e-05,  2.1532e-08,  5.3443e-05, -3.2688e-20, -3.4725e-05,\n","         1.2427e-20, -6.3211e-05, -5.7376e-06,  3.5399e-06,  2.2403e-13,\n","        -1.4392e-05, -1.4648e-04,  2.0889e-08,  2.7814e-04,  2.9814e-05,\n","         2.1840e-18, -2.1547e-05, -2.4644e-15,  1.7274e-05,  1.8260e-06,\n","         1.1725e-04, -3.4089e-05, -2.0528e-05, -4.4205e-05,  1.8377e-15,\n","         2.1341e-05,  2.6158e-06, -2.8923e-06,  5.8479e-05,  4.9366e-07,\n","         1.6009e-04,  4.7610e-06,  1.1072e-10, -1.4829e-08, -1.3902e-04,\n","         2.7765e-05, -9.1079e-08,  1.2620e-04, -1.1843e-21,  8.3904e-06,\n","         1.2758e-08,  1.6658e-04, -6.8400e-06,  2.0741e-12, -4.5411e-06,\n","         7.7712e-06, -1.4036e-06, -4.1003e-05,  1.0972e-04, -3.0174e-05,\n","         1.8707e-05, -4.5537e-05, -2.3489e-04,  1.6890e-04,  2.3396e-05,\n","         2.7030e-16,  3.7845e-13, -2.8067e-06, -3.4500e-05,  9.5017e-24,\n","        -4.3285e-05, -1.0759e-07, -1.8367e-05,  1.2494e-05,  6.2910e-05,\n","        -1.1758e-09, -2.4554e-04, -4.4733e-25, -2.6427e-06,  2.0431e-05,\n","        -2.0292e-16, -3.5146e-05,  1.4220e-05, -5.1161e-05,  1.0760e-04,\n","         1.1582e-04,  8.0732e-07,  4.9882e-05, -4.9414e-05, -2.8899e-04,\n","         5.9747e-06, -2.6077e-06, -4.7106e-06,  8.8185e-14, -3.7885e-06,\n","        -1.7410e-04,  5.1170e-05,  2.4516e-05, -3.9109e-05,  3.7137e-05,\n","        -6.8777e-07, -1.2324e-04,  1.7694e-05,  5.6791e-12,  2.4035e-05,\n","        -1.9148e-05,  8.5998e-13,  3.0977e-24,  1.4333e-05,  1.3747e-05,\n","         9.8065e-05,  8.4297e-06,  7.1034e-05, -3.3488e-17,  3.5917e-16,\n","        -6.3362e-05, -2.6015e-06, -4.7385e-05, -1.1405e-09,  3.7498e-05,\n","        -3.4019e-05,  5.9491e-07, -2.5712e-10,  2.9543e-07,  1.0149e-05,\n","         3.7950e-13,  7.7466e-11, -1.0020e-05,  5.4186e-10,  5.0074e-05,\n","        -3.3048e-06,  2.0401e-04,  9.9596e-05, -5.3014e-05,  1.4159e-06,\n","         5.7034e-05,  8.6663e-09,  3.6288e-09,  5.1756e-07, -1.5530e-05,\n","        -6.2890e-11,  2.4838e-05,  6.6880e-06, -2.3434e-05, -7.9224e-05,\n","        -1.2121e-06,  1.3849e-05,  5.6693e-06, -1.2482e-07,  3.3035e-06,\n","         6.3655e-13, -1.4092e-06,  7.4677e-14, -1.0624e-04, -6.3595e-10,\n","         2.0946e-06,  2.8765e-09,  3.7925e-04,  4.8009e-05, -2.7986e-10,\n","        -8.6595e-07, -3.5178e-05,  1.1826e-04, -9.7265e-08, -1.9699e-24,\n","        -4.8936e-05,  9.3687e-06,  9.1900e-05,  2.8111e-12, -1.4277e-11,\n","         4.1306e-06,  2.5578e-06,  8.0901e-06,  1.1782e-04,  1.5965e-03,\n","        -3.5779e-05,  3.0524e-21, -2.0230e-23, -4.2264e-05, -3.6313e-05,\n","         8.7142e-05,  2.5011e-05,  9.5588e-05,  3.4737e-06,  7.4113e-05,\n","        -2.3172e-06, -9.2388e-05,  1.1500e-05,  8.5158e-05, -1.8939e-06,\n","         2.8903e-25,  1.7905e-05,  5.7388e-06, -2.8881e-05, -6.4728e-06,\n","         3.8193e-05, -1.0788e-04,  1.0919e-04,  1.1352e-06,  5.4779e-05,\n","         4.6021e-04, -4.0786e-05,  3.4744e-05,  6.4898e-05,  2.6151e-06,\n","         4.9494e-16,  4.5480e-12,  5.1780e-14, -5.2484e-08,  3.8253e-05,\n","         4.6094e-05,  1.3531e-04, -3.1332e-04, -4.4415e-05, -8.5472e-04,\n","         6.9699e-06,  1.6170e-18, -4.0346e-05,  1.4350e-16, -8.5487e-04,\n","        -6.0139e-11,  5.5848e-05,  7.0141e-16,  7.8437e-06,  1.1764e-05,\n","         2.4203e-17,  4.9161e-05,  3.2547e-04,  1.1045e-15,  8.2751e-26,\n","        -8.7911e-05,  3.5662e-04, -9.4132e-05,  1.0710e-06,  5.0209e-06,\n","         6.0215e-05, -1.8903e-05, -5.6264e-11,  1.4093e-07,  9.2368e-05,\n","        -6.8167e-10,  3.8101e-16,  4.6380e-05, -7.5047e-06,  6.2838e-06,\n","         2.5463e-05, -3.7740e-05,  2.7007e-04,  7.1145e-06,  1.2551e-10,\n","        -8.6020e-07,  1.1908e-24, -1.1192e-05,  9.0076e-06,  2.2642e-05,\n","        -1.8932e-06,  3.7275e-10,  4.8084e-06,  9.9269e-10,  5.9672e-05,\n","         3.7482e-09, -2.0308e-05,  1.5042e-08, -6.7625e-11, -2.3534e-05,\n","        -1.9467e-07,  7.3945e-23, -1.1488e-07, -1.3121e-05, -7.7461e-05,\n","        -5.0315e-06,  8.3298e-05, -2.2664e-04,  1.2607e-11, -1.2861e-04,\n","         2.0572e-04, -5.1770e-08, -7.8821e-05, -2.4549e-06,  2.6546e-04,\n","         1.7551e-04, -4.9746e-05, -3.1123e-09,  4.9796e-05,  4.3286e-08,\n","        -7.5061e-06, -1.5525e-05, -2.2862e-05,  8.9514e-05, -4.8504e-05,\n","         9.4761e-08, -7.6396e-05, -4.6740e-07,  6.4925e-08,  4.3598e-16,\n","         1.7309e-06,  4.0482e-05, -5.2802e-05,  1.1419e-04,  3.0663e-20,\n","         1.8728e-05, -2.6712e-05, -6.3959e-05,  4.9418e-06,  1.8430e-04,\n","         4.2665e-18, -6.4463e-05, -8.8395e-05,  9.2167e-05, -6.4999e-06,\n","        -3.2444e-04, -3.8851e-05, -4.1862e-04, -4.0738e-05,  6.2755e-06,\n","         2.0615e-14,  1.4158e-15, -2.0682e-12,  7.4445e-17,  8.1480e-13,\n","         5.6353e-20, -8.2014e-06, -4.1134e-11,  3.8538e-12,  3.1376e-05,\n","        -4.4495e-05, -5.4355e-05,  5.4139e-05, -1.0136e-04,  4.4160e-06,\n","         1.5440e-04, -8.9166e-06,  6.7754e-09, -4.0203e-11,  3.4324e-05,\n","        -5.7098e-06,  1.1946e-04, -4.4647e-17,  4.4691e-06,  9.1749e-06,\n","        -6.3042e-06, -4.7454e-05, -1.3602e-10, -1.0082e-04,  9.8291e-06,\n","         2.0385e-08,  3.3511e-04,  3.0302e-05, -2.8471e-05,  1.4269e-04,\n","        -4.0530e-05, -1.4339e-05,  2.0823e-18,  1.4217e-06,  1.0772e-05,\n","         1.0238e-06, -8.3353e-05, -1.2463e-11,  3.4997e-05,  1.0867e-22,\n","        -3.2409e-05, -1.9054e-04,  2.2289e-05,  6.8341e-05,  4.4827e-10,\n","        -5.0157e-06, -1.9729e-05,  3.6918e-21,  1.1165e-05,  5.6854e-22,\n","        -7.7380e-07,  1.7334e-05,  4.0304e-16, -9.8067e-05,  2.1777e-05,\n","         1.1087e-16, -4.3816e-17,  1.5890e-05, -8.7066e-04,  2.2150e-22,\n","         1.2871e-04,  6.9367e-06,  1.9435e-05,  3.8948e-07,  1.3448e-22,\n","         3.2668e-18, -1.9214e-05, -7.0736e-05, -4.2263e-07,  6.2798e-05,\n","         3.2217e-05,  1.1673e-04,  5.6375e-08, -5.1549e-06, -5.9626e-06,\n","        -4.3123e-06, -1.3701e-05, -2.6468e-06, -1.8159e-06,  5.6614e-13,\n","         9.1591e-06, -4.9275e-06, -3.4730e-05, -4.6911e-07, -2.6873e-05,\n","        -3.3061e-04,  3.7571e-06, -9.0010e-06, -4.2508e-11, -2.9471e-05,\n","         3.0244e-07, -6.1264e-05,  1.0921e-03, -4.7992e-06,  4.3546e-05,\n","         1.7356e-12, -2.2869e-06, -1.2634e-06, -1.0551e-04, -2.0782e-04,\n","         2.0502e-05, -6.7936e-06,  9.0560e-05,  2.1694e-05, -2.3130e-05,\n","         1.9956e-06,  1.0447e-04,  2.3810e-06, -7.2532e-05, -9.5338e-17,\n","        -3.9374e-07,  9.4384e-12,  2.0806e-11, -5.0816e-06,  1.8528e-26,\n","         8.0769e-07,  1.7743e-07,  9.6895e-14,  3.0041e-10,  1.9290e-13,\n","         6.8558e-22, -1.2867e-05,  1.6757e-26, -1.2499e-04, -7.2711e-11,\n","         7.1634e-06,  1.1837e-04,  1.7363e-04, -1.1055e-04, -8.8865e-06,\n","        -6.6356e-05,  1.1743e-05, -8.3221e-06, -9.6684e-06,  1.4632e-25,\n","         5.0450e-04, -5.6967e-05, -7.9425e-05, -4.1453e-05, -4.0009e-04,\n","        -2.7289e-04,  3.8769e-06, -6.2533e-06, -1.9172e-05, -1.9009e-05,\n","         1.0315e-11, -3.8653e-07, -6.1502e-05, -6.7877e-06, -2.2009e-10,\n","         9.2026e-22, -1.6702e-06, -1.0967e-05, -3.1009e-04,  1.3599e-07,\n","        -1.0199e-07, -1.3392e-04,  5.0272e-21,  1.7594e-19, -1.2230e-04,\n","        -1.1788e-04, -7.5619e-05, -2.8386e-05,  5.7505e-05, -1.0022e-05,\n","        -2.0147e-04,  3.1630e-04,  9.6210e-06,  1.1496e-04,  1.4913e-05,\n","         1.1462e-19, -5.1484e-06, -6.7682e-05,  3.3947e-05, -1.0242e-05,\n","        -2.1904e-06, -3.5746e-08, -1.4129e-04,  6.2646e-26, -7.6650e-10,\n","         1.1436e-04,  8.3141e-05,  3.3456e-04,  8.2342e-05, -4.8175e-05,\n","         1.8242e-28,  2.4978e-05,  6.0147e-14,  3.5648e-04, -2.7232e-05,\n","         2.0613e-05,  5.6947e-07,  8.5737e-16,  6.8661e-07, -5.9042e-07,\n","        -2.3516e-05, -4.8557e-05, -7.5435e-10,  2.6213e-04,  1.3489e-26,\n","        -7.2981e-11,  4.8515e-13, -2.7742e-02, -8.5859e-09,  8.3077e-16,\n","         1.4042e-08,  1.5623e-05,  1.1332e-05, -3.9718e-06,  4.5545e-08,\n","        -6.1516e-05, -1.8407e-04,  1.0499e-23, -6.4716e-11, -5.1233e-05,\n","        -4.8027e-05,  4.6846e-10, -3.6431e-05,  7.8664e-15, -2.4114e-05,\n","         1.5784e-05,  5.9143e-19,  4.2443e-17,  2.4055e-05,  2.6943e-05,\n","         1.1388e-04, -4.1813e-05,  2.0035e-07, -2.7344e-07,  1.2771e-04,\n","        -1.4604e-04, -3.1485e-04,  3.9930e-05,  3.0618e-06,  2.4328e-05,\n","         5.6703e-25, -1.1166e-05,  3.7701e-06,  7.0259e-06,  4.0710e-05,\n","        -3.2512e-11,  4.7261e-05,  2.5639e-05, -1.1230e-09,  4.1930e-09,\n","         3.8623e-16,  9.5667e-13,  3.3458e-04, -3.6343e-09,  9.2708e-17,\n","         1.3345e-05, -8.4024e-06,  1.7333e-07,  1.7778e-16, -1.0950e-08,\n","        -5.7095e-09,  1.5115e-06, -1.2863e-04,  1.0710e-04,  4.9689e-06,\n","         1.6990e-10,  3.1316e-06,  2.0903e-06,  5.3030e-05,  3.7368e-04,\n","        -1.1558e-04, -1.4365e-04, -1.6165e-05, -1.5337e-04, -1.8788e-05,\n","        -2.0822e-04,  2.6221e-05,  1.1818e-17, -4.0278e-07, -2.6192e-04,\n","         1.1758e-06,  5.9758e-12,  6.8060e-05, -1.3323e-05, -2.5535e-09,\n","         2.3849e-06,  2.3483e-05, -1.4675e-04, -1.2500e-05, -1.1120e-05,\n","         7.4873e-05, -2.0621e-06, -2.0276e-05,  7.1374e-06, -7.0173e-06,\n","         4.1343e-08, -3.4884e-07, -1.5318e-05,  4.7619e-06, -1.2389e-24,\n","         2.5480e-05, -1.6680e-05,  1.8474e-24,  1.0028e-13,  2.1120e-05,\n","        -5.5253e-05,  1.1064e-05,  2.2810e-05, -4.4081e-05,  1.9710e-22,\n","        -2.8482e-04,  1.5916e-05,  1.6459e-09, -6.0888e-06,  1.8447e-20,\n","        -4.1732e-05, -1.3056e-04, -7.3391e-06, -7.7563e-06,  1.1079e-04,\n","         4.6270e-06,  2.7736e-05, -1.0398e-04,  3.2475e-06, -9.1845e-05,\n","         2.2166e-06,  2.5436e-05, -1.3808e-05,  2.1389e-04, -4.0241e-11,\n","         2.6108e-10,  1.1492e-04,  4.6945e-06,  3.6813e-06, -1.3722e-04,\n","        -2.5898e-04, -2.5255e-04,  7.6309e-05, -4.5769e-05, -8.4967e-21,\n","         2.8445e-05, -6.2731e-06,  1.1254e-07,  4.4706e-05, -1.0877e-04,\n","         4.7907e-04, -1.4875e-04,  1.0145e-18, -2.3964e-06, -6.0889e-05,\n","         4.3432e-06,  7.5602e-04,  8.4918e-27, -8.2287e-06,  2.9513e-06,\n","         1.9471e-06,  2.3880e-06,  3.1267e-04,  2.2164e-04, -1.2900e-04,\n","         3.4090e-07,  6.5013e-05,  2.0412e-04, -5.5397e-05, -2.9664e-05,\n","         4.2443e-17,  6.5789e-07,  2.8098e-05,  4.7107e-26,  2.9515e-25,\n","        -2.3693e-05, -2.6496e-06, -7.1293e-06, -4.6698e-05,  6.3270e-08,\n","        -1.1518e-23,  5.3402e-05, -3.1668e-05,  1.4921e-11,  1.4366e-06,\n","         1.1334e-06, -8.2537e-05, -6.6105e-05, -5.9777e-06, -4.3668e-07,\n","         1.9496e-04, -9.2456e-07,  1.6149e-10,  2.6206e-04, -1.9722e-06,\n","         7.0928e-06,  2.4182e-24,  3.3946e-05,  2.1491e-04,  6.4488e-06,\n","        -3.0818e-04,  1.9873e-07, -1.9359e-06,  3.0236e-05, -2.1825e-04,\n","        -2.1358e-04, -4.2874e-06,  1.3504e-19, -3.3766e-12, -6.1084e-06,\n","         7.0947e-13,  4.3812e-19,  2.2731e-06,  3.0565e-23,  4.0440e-03,\n","         2.9811e-04, -4.0300e-05,  2.9877e-05, -2.1029e-04,  8.7755e-30,\n","         1.0887e-04,  3.0093e-21, -2.2925e-06,  1.5391e-07,  2.3052e-05,\n","        -1.2353e-07,  4.3803e-05,  7.3806e-05, -3.4064e-06,  1.6346e-06,\n","         6.6144e-04,  1.0066e-04, -1.3809e-05,  1.2963e-04, -1.6290e-04,\n","        -8.6498e-05,  1.4230e-07,  1.1464e-04,  2.4093e-04,  1.9310e-22,\n","         1.6623e-04, -7.4135e-05,  4.5478e-06, -5.6979e-05,  1.5590e-19,\n","         9.8386e-06,  4.0203e-05,  7.7591e-07, -2.7398e-05,  1.3912e-04,\n","         6.1887e-13,  1.4104e-07, -1.9957e-06,  1.2908e-05,  1.6176e-05,\n","         2.9740e-05,  2.8217e-04,  1.2798e-04,  1.3829e-05,  3.8966e-12,\n","         1.1485e-04,  3.5412e-06, -4.6523e-07,  4.7448e-06, -6.2839e-06,\n","        -3.9983e-06,  9.7423e-05, -9.3057e-06, -1.3012e-04, -1.6949e-08,\n","         3.7341e-24, -6.4342e-06,  1.1653e-08,  5.5808e-04,  1.5524e-04,\n","        -1.5031e-04,  1.4816e-05,  6.2216e-06, -4.3573e-06, -2.0095e-07,\n","        -1.3069e-07,  3.3186e-27,  7.6968e-18, -1.2186e-07,  6.3344e-10,\n","        -2.0938e-05,  1.5501e-09,  4.9444e-05,  6.5975e-05, -4.6188e-06,\n","        -7.1426e-07,  9.3403e-26, -8.2436e-06,  1.6975e-06, -1.2667e-08,\n","         1.0140e-04,  2.3411e-05, -9.8692e-05, -4.1830e-05,  1.7966e-05,\n","        -3.4304e-06, -1.1781e-09, -5.4361e-05, -4.0731e-04,  2.5949e-06,\n","         7.5613e-05,  2.0077e-06, -8.8150e-06, -8.7985e-06,  2.4678e-05,\n","         3.0264e-04, -1.1677e-06,  1.4444e-07,  7.6431e-07, -7.1517e-07,\n","         1.8811e-05, -1.4384e-05,  7.2041e-13,  9.5376e-06,  8.5549e-06,\n","         4.7008e-16, -1.1397e-04, -2.6037e-22,  7.7193e-13,  2.6031e-05,\n","        -1.2434e-05,  5.1703e-05,  9.4648e-06, -1.8002e-07, -1.4694e-04,\n","        -1.5659e-04, -1.5357e-05,  1.8706e-05, -1.1281e-06,  2.0967e-10,\n","         1.9442e-04, -8.9122e-05, -6.7795e-05,  2.6333e-06,  2.6086e-05,\n","         4.3667e-05,  3.2894e-04, -6.6720e-05, -4.0471e-05,  4.7821e-07,\n","         4.3244e-04,  4.8192e-08, -1.7343e-05,  6.1834e-10, -8.0714e-10,\n","        -1.8442e-04,  1.5408e-18,  1.1678e-08,  1.8090e-08,  1.4953e-05],\n","       device='mps:0'), 'exp_avg_sq': tensor([4.1333e-07, 1.6012e-06, 6.9053e-07, 4.8954e-07, 1.1245e-06, 9.2825e-07,\n","        5.7040e-07, 6.0681e-07, 9.1556e-07, 5.6040e-07, 8.3633e-07, 8.2647e-07,\n","        1.8617e-06, 1.5486e-06, 9.5017e-07, 7.5996e-06, 9.3737e-07, 9.8672e-07,\n","        7.1607e-07, 1.5437e-06, 7.1063e-07, 1.0355e-06, 7.4780e-07, 9.7943e-07,\n","        7.5978e-07, 1.4328e-06, 8.1369e-07, 6.5760e-07, 1.0563e-06, 8.6186e-07,\n","        1.2241e-06, 8.6457e-07, 9.3279e-07, 5.6511e-07, 5.7810e-07, 8.6029e-06,\n","        4.9853e-07, 4.0520e-06, 1.6118e-06, 9.1655e-07, 9.8901e-07, 4.8217e-07,\n","        8.2883e-07, 8.2678e-07, 8.3427e-07, 1.1555e-06, 7.7333e-07, 5.1879e-07,\n","        3.7120e-05, 3.9064e-05, 9.1136e-07, 7.5674e-07, 8.5279e-07, 1.4944e-06,\n","        8.0507e-07, 5.9538e-07, 1.0320e-06, 9.0154e-07, 8.9765e-07, 7.4784e-07,\n","        1.0699e-06, 6.3696e-07, 6.1418e-07, 7.8607e-07, 7.1974e-07, 7.8465e-07,\n","        5.4207e-07, 7.6453e-07, 7.6453e-07, 8.7508e-07, 1.0677e-06, 1.2706e-06,\n","        3.5227e-06, 7.8511e-07, 8.8661e-07, 2.6878e-06, 1.3648e-06, 6.5474e-07,\n","        9.0697e-07, 6.1953e-07, 1.5194e-06, 8.9439e-07, 1.9811e-06, 7.0309e-07,\n","        1.2114e-06, 1.2061e-06, 3.6940e-06, 6.9360e-07, 4.1745e-07, 2.5581e-06,\n","        9.2593e-07, 2.3459e-06, 7.4075e-07, 7.5154e-07, 7.4619e-07, 6.7760e-07,\n","        1.2458e-06, 7.9597e-07, 1.0821e-06, 9.4695e-07, 1.9173e-06, 1.1288e-06,\n","        8.7168e-07, 6.2064e-07, 6.6764e-07, 9.4706e-07, 8.3106e-07, 6.4711e-07,\n","        9.0091e-07, 1.0686e-06, 7.0091e-07, 4.8250e-07, 9.7346e-07, 8.7920e-07,\n","        5.0041e-04, 9.2026e-07, 1.1867e-05, 9.8514e-07, 7.5871e-07, 7.0013e-07,\n","        5.8814e-07, 8.1029e-07, 1.0588e-06, 9.1470e-07, 7.0820e-07, 9.3771e-07,\n","        5.7056e-07, 1.0267e-06, 6.2173e-07, 9.9395e-07, 6.6081e-07, 1.3249e-06,\n","        1.1086e-06, 1.4696e-06, 1.8013e-06, 4.4792e-07, 7.8876e-07, 1.3431e-06,\n","        1.0338e-06, 4.9035e-07, 6.5496e-07, 1.1490e-06, 8.4963e-07, 8.2128e-07,\n","        6.8139e-07, 6.8241e-07, 1.0892e-06, 8.6977e-07, 8.6033e-07, 4.5996e-07,\n","        1.3539e-06, 9.2136e-07, 5.1505e-07, 9.7726e-07, 6.2854e-06, 6.4509e-07,\n","        3.9305e-07, 6.7380e-07, 7.3327e-06, 6.0325e-07, 1.1445e-06, 8.0526e-07,\n","        7.0050e-07, 5.0451e-07, 9.5047e-07, 1.0868e-06, 9.5490e-07, 8.5628e-07,\n","        7.3998e-07, 6.9038e-07, 8.8430e-07, 6.8972e-07, 1.6769e-06, 5.6459e-07,\n","        5.5583e-07, 7.8441e-07, 7.2783e-07, 1.0634e-06, 4.9426e-07, 7.2680e-07,\n","        5.5716e-07, 7.2430e-07, 8.0077e-07, 7.8625e-07, 1.0005e-06, 9.8144e-07,\n","        7.2188e-07, 3.8968e-06, 9.1237e-07, 1.0503e-06, 1.6127e-06, 7.0562e-07,\n","        6.9729e-07, 7.6587e-07, 1.1689e-06, 1.6447e-06, 7.7527e-07, 8.9604e-07,\n","        6.0888e-07, 1.9366e-05, 1.1436e-06, 6.7585e-07, 7.5662e-07, 9.5087e-07,\n","        1.1247e-06, 1.1147e-06, 4.2076e-07, 1.0304e-06, 4.9161e-07, 1.3325e-06,\n","        1.2826e-06, 1.0583e-06, 1.0419e-06, 8.1227e-07, 1.0101e-06, 6.7846e-07,\n","        1.1389e-06, 1.1478e-06, 7.9071e-07, 7.8509e-07, 7.5267e-07, 9.9278e-07,\n","        2.2599e-06, 3.8461e-07, 1.0992e-06, 1.1867e-06, 7.4064e-07, 4.9035e-07,\n","        7.9958e-07, 1.1429e-06, 1.2998e-06, 1.2250e-06, 1.4179e-05, 1.1480e-06,\n","        5.5328e-07, 5.9672e-07, 1.3719e-06, 2.2515e-06, 8.6888e-07, 5.8866e-07,\n","        1.0377e-06, 1.2380e-06, 2.6935e-04, 8.3826e-07, 1.0252e-06, 5.6448e-07,\n","        5.2256e-07, 7.3594e-07, 1.1071e-06, 9.1319e-07, 4.9792e-07, 3.8282e-07,\n","        6.5679e-07, 5.0573e-07, 6.4752e-07, 7.6586e-07, 3.9808e-07, 5.2777e-07,\n","        7.3813e-07, 9.0359e-07, 6.6313e-07, 6.4710e-07, 9.5844e-07, 1.0519e-06,\n","        6.1850e-07, 5.9777e-07, 6.6580e-07, 9.0256e-07, 8.4137e-07, 1.1228e-06,\n","        7.9675e-07, 9.1108e-07, 9.5923e-07, 5.9362e-07, 7.6505e-07, 7.2384e-07,\n","        5.6408e-07, 5.8543e-07, 1.2337e-06, 8.8274e-07, 1.0739e-06, 8.8313e-07,\n","        8.1005e-07, 1.2464e-06, 1.1517e-06, 9.1062e-07, 8.8637e-07, 1.0388e-06,\n","        4.8939e-06, 1.5900e-05, 7.1567e-07, 5.9170e-07, 1.5440e-06, 6.7994e-07,\n","        8.8663e-07, 6.8269e-07, 5.0962e-06, 1.4575e-06, 9.4394e-07, 6.7887e-07,\n","        9.2639e-07, 5.2612e-07, 5.2797e-07, 7.1672e-07, 7.8074e-07, 9.0433e-07,\n","        8.9017e-07, 6.8657e-07, 2.2549e-06, 7.1457e-07, 8.9318e-07, 1.6847e-06,\n","        8.6539e-07, 9.8556e-07, 7.6049e-07, 9.4776e-07, 1.1283e-06, 4.7392e-06,\n","        1.2512e-06, 7.9673e-07, 7.2498e-07, 8.4039e-07, 8.7553e-07, 1.9984e-06,\n","        8.7552e-07, 8.2805e-07, 5.0022e-07, 1.5804e-06, 8.5954e-07, 5.8772e-07,\n","        5.3864e-07, 1.2842e-06, 6.5580e-07, 6.7121e-07, 9.2280e-07, 1.2581e-06,\n","        1.1735e-06, 1.2760e-06, 9.0650e-07, 4.6112e-07, 7.5947e-07, 7.6593e-07,\n","        9.1511e-07, 1.1869e-06, 9.2374e-07, 7.1974e-07, 7.1323e-07, 2.2765e-06,\n","        9.1373e-07, 1.1552e-06, 8.7506e-07, 1.5235e-06, 7.6865e-07, 1.1026e-06,\n","        1.2875e-06, 6.0670e-07, 6.8290e-07, 6.9916e-07, 5.5804e-07, 8.7920e-07,\n","        1.3210e-06, 6.6703e-07, 7.3193e-07, 6.2045e-07, 9.0898e-07, 7.2652e-07,\n","        6.6098e-07, 7.8653e-07, 9.8426e-07, 8.0957e-07, 9.1229e-07, 1.2020e-06,\n","        8.1035e-07, 7.6022e-07, 1.0001e-06, 4.7176e-07, 9.1298e-07, 1.1690e-06,\n","        7.2858e-07, 8.5636e-07, 1.0921e-06, 6.6523e-07, 1.0443e-06, 2.2488e-05,\n","        9.3868e-07, 8.8891e-07, 1.5373e-06, 1.0363e-06, 5.3253e-07, 6.5018e-07,\n","        5.8023e-07, 6.2431e-07, 8.0087e-07, 8.4034e-07, 5.7913e-07, 6.5144e-07,\n","        7.3859e-07, 1.0064e-06, 5.5447e-07, 5.0782e-07, 1.1033e-06, 4.3316e-06,\n","        7.2549e-07, 5.9441e-07, 9.1482e-07, 8.4947e-07, 8.6705e-07, 1.0528e-06,\n","        1.1168e-06, 1.5476e-06, 4.9069e-07, 8.1228e-07, 1.1029e-06, 5.0816e-07,\n","        8.4071e-07, 7.3884e-07, 6.9364e-07, 7.5215e-07, 4.5188e-07, 7.5437e-07,\n","        1.1565e-06, 1.0686e-06, 1.0751e-06, 1.2212e-06, 6.4535e-07, 1.0734e-06,\n","        2.5805e-06, 6.8187e-07, 7.3759e-07, 1.0196e-06, 9.4332e-07, 6.9436e-07,\n","        1.9510e-06, 8.2691e-07, 1.2659e-06, 6.0882e-07, 8.9651e-07, 1.2474e-06,\n","        5.8199e-07, 6.4020e-07, 8.9224e-07, 1.6371e-06, 1.4287e-06, 7.7378e-07,\n","        1.8446e-06, 1.0049e-06, 7.2516e-07, 1.1943e-06, 8.8864e-07, 6.8079e-07,\n","        1.1940e-06, 1.0857e-06, 8.5188e-07, 9.0686e-07, 7.0405e-07, 5.3151e-07,\n","        1.4346e-06, 5.6273e-07, 7.7493e-07, 5.5874e-07, 1.0034e-06, 9.8901e-07,\n","        7.8666e-07, 1.1897e-06, 1.1670e-06, 5.6096e-07, 6.8408e-07, 4.6335e-07,\n","        6.8141e-07, 2.7217e-06, 9.9391e-07, 9.7645e-07, 9.2644e-07, 6.3261e-07,\n","        6.9685e-07, 7.6304e-07, 6.4050e-07, 8.4392e-07, 5.7909e-07, 5.2815e-07,\n","        5.9495e-07, 1.3894e-06, 1.6911e-06, 8.9874e-07, 5.8194e-07, 9.8779e-06,\n","        9.0459e-07, 7.5624e-07, 6.9995e-07, 5.9842e-07, 5.9926e-07, 6.3529e-07,\n","        9.8830e-07, 7.1187e-07, 3.2399e-04, 1.2311e-06, 7.3967e-07, 9.9661e-07,\n","        8.6199e-07, 1.0990e-06, 5.9100e-07, 6.2781e-07, 1.4840e-06, 4.7271e-07,\n","        7.8033e-07, 1.0159e-06, 7.7102e-07, 5.7465e-07, 1.1116e-06, 6.6936e-07,\n","        8.2767e-07, 7.1760e-07, 1.1703e-06, 1.7536e-06, 1.0124e-06, 8.6539e-07,\n","        7.6110e-07, 7.7557e-07, 5.9242e-07, 1.0973e-05, 8.7617e-07, 5.3731e-07,\n","        1.0266e-06, 7.1716e-07, 8.4294e-07, 1.7115e-06, 7.4568e-07, 8.2850e-07,\n","        8.9862e-07, 8.1817e-07, 5.7591e-07, 6.8099e-07, 6.4180e-07, 7.6365e-07,\n","        1.0056e-06, 6.4719e-06, 7.7220e-07, 9.3098e-07, 6.5745e-07, 6.0739e-07,\n","        9.1524e-07, 1.0111e-06, 9.5421e-07, 6.6545e-07, 8.1122e-07, 9.4865e-07,\n","        1.3734e-06, 8.1456e-07, 7.1543e-07, 7.4979e-07, 1.1617e-06, 9.8401e-07,\n","        1.0751e-06, 1.1275e-06, 8.7195e-07, 9.8081e-07, 6.0049e-07, 7.9256e-07,\n","        7.2502e-07, 1.7326e-06, 7.2002e-07, 8.6431e-07, 1.2101e-06, 7.1774e-07,\n","        8.7360e-07, 8.6297e-07, 7.4852e-07, 9.8460e-07, 7.6971e-07, 1.7377e-05,\n","        6.1624e-07, 6.7314e-07, 9.9480e-07, 1.3206e-06, 1.5789e-06, 5.0531e-07,\n","        1.1207e-06, 1.4242e-06, 1.1187e-06, 3.3656e-07, 6.3368e-07, 1.4014e-06,\n","        9.2506e-07, 1.2334e-06, 4.8172e-07, 8.8086e-07, 6.5962e-07, 4.2804e-07,\n","        1.0077e-06, 5.0412e-07, 1.1684e-06, 1.0545e-06, 4.5955e-07, 4.4257e-07,\n","        9.5922e-07, 8.8832e-07, 8.1391e-07, 9.6161e-07, 7.4399e-07, 9.9507e-07,\n","        4.2937e-07, 5.7102e-07, 5.1167e-07, 1.1576e-06, 6.4857e-07, 6.8021e-07,\n","        9.9214e-07, 7.9827e-07, 5.9006e-07, 7.7057e-07, 7.8433e-07, 6.9712e-07,\n","        9.2494e-06, 1.6244e-06, 1.5122e-06, 1.3754e-06, 1.1948e-06, 1.2454e-06,\n","        9.4316e-07, 6.5626e-06, 1.0026e-06, 9.1685e-07, 1.5847e-06, 9.7693e-07,\n","        6.9567e-07, 9.3067e-07, 1.1446e-06, 9.7863e-07, 2.3283e-06, 5.0286e-07,\n","        1.1202e-06, 3.3714e-05, 5.1831e-07, 7.0844e-07, 6.0756e-07, 1.2175e-05,\n","        9.8015e-07, 1.1503e-06, 1.2844e-06, 6.5438e-07, 3.0826e-06, 6.0900e-07,\n","        1.6263e-06, 1.2582e-06, 6.8258e-07, 1.0170e-05, 4.7890e-07, 1.1783e-06,\n","        1.0557e-06, 1.1803e-06, 1.2253e-06, 9.2596e-07, 7.4869e-07, 9.5791e-07,\n","        7.8756e-07, 5.4210e-07, 9.0254e-07, 1.4637e-04, 6.7523e-07, 1.0597e-06,\n","        5.8624e-07, 6.6617e-07, 1.4546e-06, 7.0231e-07, 8.0023e-07, 7.7186e-07,\n","        8.8244e-07, 1.6557e-06, 8.0690e-07, 1.0543e-06, 4.9150e-07, 6.9977e-07,\n","        6.3581e-07, 6.3461e-07, 1.1517e-06, 1.0515e-06, 1.0254e-06, 8.5249e-07,\n","        6.8558e-07, 8.7707e-07, 5.3539e-07, 9.0691e-07, 1.0533e-06, 9.0298e-07,\n","        1.0857e-06, 6.9370e-07, 1.3794e-06, 1.6149e-06, 1.0475e-06, 1.1983e-06,\n","        7.8845e-07, 1.5597e-06, 8.4403e-07, 7.7898e-07, 1.4943e-06, 5.7633e-07,\n","        8.0450e-07, 7.2387e-07, 5.5138e-07, 8.9200e-07, 1.8023e-05, 1.0662e-06,\n","        1.1393e-06, 9.6724e-07, 9.4615e-07, 6.5412e-07, 1.0016e-06, 5.3508e-07,\n","        6.8283e-07, 7.7889e-07, 7.5173e-07, 1.9188e-06, 7.9992e-07, 9.2702e-07,\n","        6.4427e-07, 1.2624e-06, 5.9090e-07, 5.5479e-07, 8.0403e-07, 1.5772e-06,\n","        1.3375e-06, 5.2843e-07, 5.9440e-07, 8.0478e-07, 8.2665e-07, 8.8412e-07,\n","        6.6193e-07, 3.8857e-06, 6.0535e-07, 8.9559e-07, 1.2029e-06, 5.7562e-07,\n","        8.9352e-07, 3.2116e-05, 1.2328e-06, 1.3347e-06, 6.5502e-07, 7.2432e-07,\n","        8.4167e-07, 1.1037e-06, 7.3117e-07, 7.7201e-06, 6.1453e-07, 4.5992e-07,\n","        9.0942e-07, 3.5721e-06, 6.0854e-07, 6.7623e-07, 7.0446e-07, 9.9289e-07,\n","        8.8560e-07, 6.9787e-07, 1.3006e-06, 9.9014e-07, 9.6589e-07, 5.4997e-07,\n","        7.2102e-07, 7.9027e-07, 8.9595e-07, 1.5498e-06, 1.1166e-06, 1.1018e-06,\n","        5.3910e-07, 7.3043e-07, 8.5058e-07, 6.5327e-07, 8.4828e-07, 7.1319e-07,\n","        5.7233e-07, 6.1503e-07, 6.5567e-07, 8.9137e-07, 9.0982e-07, 1.0894e-06,\n","        1.3636e-06, 8.6145e-07, 8.1198e-07, 7.6550e-07, 8.7782e-07, 7.6156e-07,\n","        6.4349e-07, 1.2194e-06, 7.3728e-07, 7.7197e-07, 8.6374e-07, 8.2886e-07,\n","        6.9993e-07, 7.3856e-07, 8.6201e-07, 2.1164e-06, 7.7570e-07, 6.2165e-07,\n","        8.9626e-07, 1.1386e-06, 5.0986e-07, 1.0716e-06, 9.8531e-07, 8.5272e-07,\n","        7.1873e-07, 8.9649e-07, 4.1617e-07, 7.5104e-07, 6.2265e-07, 1.0335e-06,\n","        3.7715e-07, 7.9993e-07, 9.2613e-07, 1.0165e-06, 8.4826e-07, 9.7790e-07,\n","        5.0462e-07, 8.9891e-07, 8.6867e-07, 4.6525e-07, 1.0045e-06, 1.1505e-06,\n","        7.7328e-07, 8.4496e-07, 9.2147e-07, 1.4321e-06, 4.6604e-07, 5.0329e-07,\n","        7.1401e-07, 9.7445e-07, 5.8918e-07, 7.5495e-07, 4.5200e-06, 7.0933e-07,\n","        6.6467e-07, 1.1244e-06, 7.8863e-07, 6.2563e-07, 9.8207e-07, 2.4595e-06,\n","        5.5550e-07, 9.6719e-07, 5.1664e-07, 1.1030e-06, 5.9537e-07, 1.1439e-06,\n","        4.8526e-07, 9.0997e-07, 8.8192e-07, 1.0515e-06, 1.1168e-06, 1.9770e-06,\n","        9.3398e-07, 6.1268e-07, 9.6064e-07, 1.7388e-06, 9.4326e-07, 8.3533e-07,\n","        1.0304e-06, 8.1834e-07, 4.5039e-07, 6.9021e-07, 9.3166e-07, 1.4354e-05,\n","        2.1667e-06, 2.1275e-05, 8.4655e-07, 1.6362e-06, 9.5287e-07, 1.9291e-06,\n","        6.4203e-07, 1.0840e-06, 8.7628e-07, 9.2735e-07, 1.4791e-06, 6.7480e-07,\n","        9.1032e-07, 1.1692e-06, 6.8664e-07, 8.6842e-07, 5.5517e-07, 7.3604e-07,\n","        1.1574e-06, 9.7851e-07, 9.6046e-07, 6.3131e-07, 1.1705e-06, 6.0166e-07,\n","        9.5142e-07, 8.3938e-07, 6.9730e-07, 5.7922e-07, 1.5015e-06, 6.3624e-07,\n","        1.2478e-06, 2.0770e-06, 5.7074e-07, 7.7852e-07, 6.0443e-07, 7.2850e-07,\n","        1.0683e-06, 1.5522e-06, 1.0319e-06, 1.0273e-06, 6.4266e-07, 8.2027e-07,\n","        7.7246e-07, 1.4072e-06, 7.9766e-07, 9.6264e-07, 7.5727e-07, 6.9250e-07,\n","        9.9337e-07, 7.5420e-07, 7.7516e-07, 9.1899e-07, 6.5549e-07, 1.0049e-06,\n","        4.7981e-07, 1.1176e-06, 9.7373e-06, 6.4323e-07, 5.0868e-07, 7.7718e-07,\n","        1.1868e-06, 6.3508e-07, 1.4057e-06, 1.0887e-06, 4.7362e-07, 1.2536e-06,\n","        7.5317e-07, 8.8378e-07, 7.2505e-07, 9.1887e-07, 6.7978e-07, 5.5597e-07,\n","        5.3536e-07, 7.9339e-07, 5.5991e-07, 1.2986e-06, 8.8225e-07, 1.3299e-06,\n","        9.5379e-07, 1.0265e-06, 1.3248e-06, 2.5430e-06, 1.0086e-06, 1.2748e-06,\n","        1.0102e-06, 6.6818e-07, 8.6648e-07, 9.7868e-07, 1.5930e-06, 1.3152e-06,\n","        1.2362e-06, 5.9163e-07, 1.2712e-06, 7.0389e-07, 8.7527e-07, 1.0742e-06,\n","        9.4269e-07, 9.7942e-07, 6.4588e-07, 6.7555e-07, 6.0221e-07, 7.2729e-07,\n","        6.5218e-07, 7.6396e-07, 1.0236e-06, 5.4091e-07, 1.3550e-06, 1.4665e-06,\n","        1.0642e-06, 5.2606e-07, 1.0572e-06, 8.8661e-07, 1.0665e-06, 1.2151e-06,\n","        8.7850e-07, 5.6503e-07, 3.6836e-06, 1.0601e-06, 7.1219e-07, 1.2848e-06,\n","        6.2893e-07, 8.9155e-07, 8.6322e-07, 1.0525e-06, 6.9353e-07, 5.6940e-07,\n","        6.6207e-07, 1.7193e-06, 8.0207e-07, 7.1645e-07, 5.5378e-07, 8.1278e-07,\n","        5.3721e-07, 1.8539e-06, 7.8676e-07, 7.8956e-07, 4.7149e-07, 8.7641e-07,\n","        9.8954e-07, 1.4698e-06, 1.0356e-06, 4.3424e-07], device='mps:0')}, 4: {'step': tensor(1025.), 'exp_avg': tensor([-1.0111e-04, -2.9631e-10, -6.1335e-04, -6.2773e-05,  1.9114e-21,\n","        -5.9341e-08,  3.2259e-16, -4.7180e-04,  4.0728e-23, -5.8495e-04,\n","        -5.3509e-10, -9.0109e-04, -7.5636e-04, -9.7282e-04, -6.4208e-08,\n","        -1.2164e-04,  2.5833e-04,  6.3083e-04,  6.6456e-04,  6.9456e-04,\n","         9.0930e-05, -2.0559e-07, -2.3660e-03, -1.1713e-03, -1.8247e-05,\n","        -3.4197e-04, -2.3923e-17,  1.1909e-22,  6.2572e-22,  9.9486e-04,\n","        -8.3368e-04, -5.9937e-04, -1.4762e-04, -8.3385e-04,  1.7736e-04,\n","        -1.7520e-04,  7.8052e-14, -2.6459e-04, -2.4900e-17, -1.6104e-07,\n","        -1.6972e-09,  5.5150e-04, -2.3164e-04, -5.0014e-04, -4.2851e-04,\n","         4.6928e-04, -7.5595e-04, -7.2974e-04, -4.9148e-30,  6.9308e-05,\n","        -1.6028e-13, -5.0241e-04, -5.2754e-04, -5.3563e-07, -4.9433e-04,\n","        -7.5161e-04,  8.1507e-04,  4.5010e-10,  3.2708e-06,  3.1225e-05,\n","        -4.2549e-05, -3.4220e-04, -1.0077e-20, -6.2250e-04,  6.1723e-06,\n","        -5.3289e-04,  4.1645e-04, -7.6488e-14, -1.1687e-03, -9.2853e-22,\n","         8.4205e-04,  5.4810e-04,  5.4488e-17,  3.4175e-04, -9.2638e-07,\n","         6.8249e-04, -9.9097e-04,  7.9515e-04,  4.8703e-04, -1.0396e-03,\n","        -9.4578e-04, -8.1749e-04,  9.8045e-04, -1.1266e-08, -8.6460e-20,\n","        -4.6524e-04,  1.2965e-22,  3.1494e-26, -1.5505e-04,  1.0392e-04,\n","        -4.4609e-04,  2.7899e-11, -5.6729e-04, -1.1529e-03, -9.0110e-04,\n","         1.4277e-03,  7.0786e-04, -5.6533e-04, -7.8308e-04, -1.7564e-14,\n","        -5.8124e-04,  3.4419e-04,  9.7638e-22, -8.1727e-04,  3.4191e-04,\n","         3.9887e-14,  2.9684e-04, -7.7322e-06, -2.4962e-06,  4.6809e-05,\n","        -6.9760e-04,  3.2611e-12, -7.1114e-05, -8.0534e-04, -8.4032e-04,\n","        -7.0678e-04,  2.4193e-06,  2.2501e-04,  1.4429e-18, -4.8338e-04,\n","         3.1275e-05,  3.0722e-04,  3.6408e-05,  9.9294e-14, -1.5009e-04,\n","        -6.3183e-05,  5.5219e-04,  7.2504e-04, -5.9313e-05, -1.5375e-03,\n","        -4.3408e-04, -1.3724e-04, -7.6884e-04, -5.4019e-04,  3.8953e-04,\n","        -3.5107e-04, -6.9586e-05,  3.7415e-16,  2.0551e-09,  9.7468e-04,\n","        -8.1464e-05,  1.4435e-03,  1.8874e-04, -5.0274e-04, -6.4047e-04,\n","        -7.8355e-04,  2.4205e-08, -1.1875e-03, -1.7749e-19, -1.0056e-04,\n","         1.9667e-21, -7.9091e-05,  1.0182e-03, -9.5485e-05,  5.2973e-16,\n","        -3.1035e-04, -6.6433e-05,  1.3623e-08,  9.9966e-07,  1.0614e-04,\n","        -5.8690e-20,  1.5370e-04, -2.5431e-14, -2.9404e-04, -9.1656e-04,\n","        -5.8497e-04, -1.0575e-03,  2.0845e-04, -1.3587e-04,  5.2133e-16,\n","        -1.4847e-04,  2.3652e-04,  3.6224e-04, -3.5665e-04,  2.0804e-04,\n","        -1.2058e-03, -2.4672e-04, -7.4135e-13,  3.3024e-04, -8.3821e-06,\n","        -6.9887e-04, -5.2562e-06, -7.5002e-05, -4.0763e-21,  1.5117e-04,\n","         6.0087e-09, -1.1213e-03, -1.2704e-03, -9.0980e-13,  2.6037e-04,\n","        -1.0996e-03, -9.5138e-04, -2.5121e-04, -1.7487e-03,  1.1207e-03,\n","        -6.0225e-06,  1.1543e-04,  8.5061e-04, -2.0240e-04, -9.5823e-04,\n","        -1.4195e-14,  4.9883e-14, -3.9034e-04,  2.6159e-04,  7.2620e-24,\n","         3.3912e-04, -5.1902e-04,  5.6297e-04, -7.3585e-04,  3.5121e-05,\n","        -2.1781e-07, -3.9332e-04, -1.3996e-24, -1.0865e-03,  9.6626e-04,\n","        -1.0564e-14,  1.7188e-03, -1.0062e-03,  2.7972e-04,  4.6085e-04,\n","         4.2929e-04, -7.2482e-04, -4.2038e-05, -6.8627e-04, -5.4050e-04,\n","        -6.6747e-04, -2.9750e-04,  1.7310e-04,  8.4777e-16, -3.8877e-04,\n","         7.6437e-04, -6.9927e-04, -7.3411e-04, -4.1294e-04, -1.3076e-03,\n","        -3.6649e-04, -1.0196e-03,  1.7047e-07,  3.7266e-14,  3.4443e-06,\n","        -4.1296e-04, -4.0834e-13,  6.6235e-26,  1.0430e-05, -6.9358e-04,\n","        -9.9260e-04,  5.0949e-04, -5.2126e-04, -5.9611e-17, -1.9409e-17,\n","        -2.5725e-04, -6.6222e-04,  1.3830e-04, -6.9343e-04,  5.5011e-04,\n","        -2.7221e-04, -6.1250e-04, -6.0218e-10, -2.8351e-05,  1.3837e-06,\n","        -1.5079e-13, -2.2414e-04, -1.3683e-04,  3.4396e-07, -6.1054e-04,\n","         5.9319e-05,  1.0206e-03,  3.8475e-04,  2.4083e-04, -1.8850e-07,\n","         4.9556e-04,  8.3445e-08, -2.8215e-09,  1.6727e-04, -6.8336e-04,\n","        -7.4933e-04,  2.8007e-04, -8.5106e-05,  1.3909e-05, -8.4580e-04,\n","        -1.9453e-04, -4.2852e-04, -6.4154e-04, -2.1630e-05, -2.8463e-04,\n","        -1.5964e-13,  1.6850e-04,  3.3422e-14,  6.7948e-04, -4.0413e-04,\n","        -2.2442e-04, -3.8227e-10, -3.7204e-06,  2.0876e-04, -7.1033e-08,\n","        -1.7295e-04, -9.2421e-04,  3.0067e-04, -4.1852e-06, -1.1944e-23,\n","         3.8462e-04, -8.4634e-04,  4.1718e-04, -5.0465e-04, -6.1043e-09,\n","         2.1735e-04,  6.5572e-07, -2.7361e-04,  2.8639e-04,  2.7780e-04,\n","         3.7905e-04, -6.4379e-22, -1.1057e-22, -3.4604e-04,  4.7383e-04,\n","         1.7891e-04, -6.3999e-04,  3.0762e-04, -1.2324e-03, -1.0200e-03,\n","         4.6710e-04,  8.7487e-04, -3.1504e-04, -3.3419e-04, -1.3613e-03,\n","         4.5358e-26,  5.0903e-07,  3.3591e-07,  4.0435e-04, -8.8666e-05,\n","        -2.6976e-04,  5.8684e-04, -1.2240e-04, -4.3832e-04,  3.6639e-04,\n","         3.4312e-05, -4.0909e-04, -4.0539e-04, -4.8913e-04, -1.6289e-04,\n","         6.4119e-17,  2.0334e-11,  1.8270e-15, -2.2515e-04, -7.8950e-06,\n","        -6.1402e-04,  5.2954e-05,  1.0669e-03, -1.1947e-03, -7.8341e-04,\n","         3.4114e-05, -4.2804e-19,  2.6128e-04,  4.6657e-17, -3.4805e-04,\n","        -3.9009e-04,  5.2220e-05, -8.2898e-17, -4.1834e-04,  3.2590e-06,\n","         5.4358e-20,  1.2558e-05,  9.7325e-04, -6.6306e-17,  1.9424e-23,\n","        -2.5416e-04,  1.1979e-04,  7.3957e-05, -2.3376e-04, -2.8729e-04,\n","         9.8686e-05, -1.1704e-04, -9.9681e-09, -5.3250e-04, -2.7727e-04,\n","         1.3826e-04, -5.1189e-17, -5.3918e-04,  2.2273e-04, -1.8803e-04,\n","        -1.1153e-03, -8.7687e-04,  2.5858e-05,  1.2304e-04, -1.2919e-10,\n","        -1.9920e-04, -3.8883e-26,  5.9928e-04, -7.3291e-04,  1.1637e-05,\n","         5.1346e-04,  7.2394e-07, -8.5211e-04,  3.1466e-09,  3.7072e-04,\n","         1.2105e-03, -1.9756e-04,  2.8665e-08, -1.8328e-04,  1.5525e-04,\n","        -3.1202e-04, -7.7219e-25, -5.8690e-04, -4.5612e-05, -3.3293e-04,\n","        -2.4177e-04, -9.0922e-04,  6.1252e-06, -5.3371e-12, -8.9155e-04,\n","        -1.1058e-03, -1.1080e-03,  9.3042e-04, -8.5159e-04, -8.5987e-04,\n","        -1.4638e-03,  1.5949e-05, -1.0313e-07, -1.3069e-05, -7.9904e-04,\n","        -1.2882e-03, -4.8143e-05,  7.8807e-04, -3.5553e-04,  2.1404e-05,\n","        -1.1055e-05, -5.0645e-04, -3.0436e-04, -3.3806e-05,  3.7260e-17,\n","        -1.0147e-06, -1.4215e-05, -1.0238e-03,  1.2650e-04, -9.5585e-21,\n","        -1.5552e-06,  1.7846e-05, -5.5120e-04, -1.1036e-04,  1.4045e-03,\n","         6.4045e-19,  1.3966e-04, -5.4794e-04, -5.3044e-04, -6.7447e-04,\n","         3.4143e-04,  3.9139e-05, -4.1632e-04,  4.0705e-05, -3.6199e-04,\n","        -2.3508e-14,  1.8414e-17, -2.2696e-08, -9.1523e-19, -6.4209e-13,\n","        -1.5531e-19, -5.3860e-04, -3.4036e-05,  4.3354e-13, -7.1848e-04,\n","         3.9476e-04, -1.1412e-03,  4.0049e-05, -2.6635e-04,  4.8648e-06,\n","        -3.7998e-04, -2.3412e-04,  7.6817e-05, -1.2568e-03, -1.2911e-04,\n","         3.1354e-04, -6.3179e-04,  6.2971e-15, -1.0511e-03, -1.4612e-04,\n","         2.1670e-04, -4.4586e-04, -3.4935e-09, -1.8069e-04, -2.4943e-05,\n","         1.2034e-04,  1.0604e-04,  3.9384e-04,  3.1215e-05, -5.0469e-05,\n","        -1.5029e-05, -1.6833e-04,  8.0108e-19,  2.8317e-04, -2.4396e-04,\n","        -1.1572e-03, -7.6997e-04, -8.6152e-04, -7.2143e-04, -9.7139e-25,\n","        -1.1340e-04, -2.1871e-04,  1.0647e-04, -7.0319e-04,  1.7145e-07,\n","        -7.1827e-04, -6.7028e-04, -2.1418e-22, -3.4011e-05,  4.9245e-22,\n","        -8.8851e-05, -4.0042e-04, -3.3329e-16,  1.6517e-04,  1.9502e-04,\n","         1.2616e-16, -1.1817e-16, -1.4896e-04, -3.9879e-04, -1.9556e-20,\n","        -8.1855e-04,  8.2457e-05, -4.4887e-04, -7.4884e-05,  3.7586e-24,\n","         3.4463e-18, -1.9044e-04,  3.1796e-04,  1.4034e-04, -8.8375e-04,\n","         9.2741e-04, -5.8268e-04, -1.7571e-05, -1.2383e-04, -1.1740e-03,\n","         4.5110e-05,  4.0286e-04,  2.5032e-04, -2.2650e-04, -9.1479e-14,\n","        -1.8539e-04,  8.2490e-04, -5.5225e-04,  6.0666e-04,  2.7173e-04,\n","        -6.5756e-04, -8.8676e-08,  6.3763e-05, -3.3301e-04,  4.1746e-04,\n","         2.2158e-04, -3.4971e-04,  5.8031e-04, -5.0877e-04, -7.3546e-04,\n","         5.1575e-14,  8.0758e-04,  4.0291e-04,  5.2434e-04, -2.5312e-04,\n","        -4.8293e-04, -3.9375e-04,  1.9950e-04, -5.4930e-04,  4.5663e-04,\n","         2.5004e-04, -1.0382e-03, -7.2928e-04, -3.2778e-04, -3.5834e-16,\n","        -4.3126e-04,  2.0902e-11, -2.6590e-12, -4.4296e-04, -2.0558e-29,\n","         7.9112e-04, -4.9932e-04, -8.9134e-15,  2.7503e-11, -1.3084e-14,\n","         1.9360e-20,  6.7179e-04, -1.1000e-26, -6.2866e-04, -3.7744e-04,\n","        -4.9452e-04, -9.5537e-04, -3.6767e-04,  9.2304e-05, -6.4959e-04,\n","         1.3393e-03, -5.4701e-04,  8.2322e-05, -2.3883e-04,  2.3320e-24,\n","         2.7475e-04, -4.6739e-04, -2.7051e-04,  1.0429e-03, -2.2514e-04,\n","        -9.3334e-04, -5.5858e-04, -5.0527e-04,  4.9739e-04, -6.6949e-04,\n","        -6.7222e-04, -5.5025e-05,  1.2706e-05, -3.4042e-04, -1.0273e-04,\n","        -2.1802e-19, -7.8553e-04, -1.8765e-04,  5.0367e-04, -2.1278e-04,\n","        -5.8123e-06,  4.1462e-05,  4.4103e-22,  2.5941e-20, -1.6998e-04,\n","         3.2323e-04,  8.5708e-04, -3.5161e-04,  1.2429e-04,  1.1823e-04,\n","        -5.2051e-04,  3.3383e-04,  1.8246e-06,  9.9665e-06,  1.1163e-04,\n","        -2.0211e-20, -1.4383e-03, -3.0085e-04, -3.1086e-04,  9.6993e-05,\n","        -1.1527e-04,  5.0123e-04, -1.2619e-03,  3.1164e-27, -3.8247e-04,\n","        -8.1600e-04,  2.1209e-04,  8.6174e-05, -3.7670e-05,  5.8715e-04,\n","         6.0264e-25, -2.0022e-04,  7.1273e-16,  1.0655e-03, -3.1059e-04,\n","         1.5974e-04, -9.1914e-04, -1.9836e-17, -3.3916e-04,  2.6030e-04,\n","         4.2105e-04,  1.0399e-03, -3.2418e-09, -8.1575e-04,  1.4621e-27,\n","        -7.1111e-04,  4.7158e-14, -1.0385e-03,  2.4996e-04,  2.0264e-13,\n","         1.3579e-03,  3.2942e-04, -5.3904e-04,  6.7235e-04, -4.9032e-08,\n","         4.6518e-05,  1.2532e-04,  2.3746e-24,  9.3531e-05, -1.0198e-04,\n","        -2.8414e-04,  1.1571e-10, -8.0981e-04,  1.7548e-15, -2.5387e-05,\n","         4.8130e-04, -5.8637e-20, -5.9472e-19,  6.7436e-04, -5.2448e-04,\n","        -1.8196e-04, -1.5302e-04,  2.9590e-04, -9.8075e-04, -7.1599e-04,\n","        -1.3685e-03,  8.4384e-04,  2.8327e-06, -3.9599e-04,  5.1954e-04,\n","        -1.3069e-26,  9.1252e-04, -3.3974e-04, -2.4632e-05, -2.0738e-04,\n","        -4.1147e-05, -2.3216e-04, -7.0104e-04, -6.8647e-04,  1.1482e-10,\n","         2.8351e-17,  4.0274e-14, -4.4694e-04, -1.4095e-06,  3.2085e-17,\n","        -6.1151e-04,  5.8915e-04,  1.1630e-03,  7.9912e-15, -6.9498e-04,\n","         6.4215e-04,  7.5864e-04, -5.2135e-04,  2.9773e-04,  2.6788e-04,\n","        -5.8929e-07,  7.8419e-05, -8.9336e-04, -1.4964e-03, -2.9155e-04,\n","        -1.2354e-04,  4.5176e-05,  2.0654e-04, -4.2377e-04, -4.1066e-05,\n","        -7.1632e-04, -3.8394e-04, -7.3460e-20,  3.0821e-06, -1.6084e-04,\n","        -2.1510e-04, -1.0226e-11,  1.7732e-04, -1.1760e-04, -3.7691e-07,\n","         7.0701e-04, -5.2760e-04, -8.1451e-04,  1.4685e-04, -7.5476e-04,\n","        -9.3037e-04,  4.0065e-04, -6.7247e-04, -2.1446e-04, -8.3199e-04,\n","        -3.7100e-11,  1.1603e-03,  1.3370e-04, -5.4032e-04, -4.2236e-24,\n","        -2.2131e-04,  4.5977e-05,  1.4965e-24,  4.2481e-14, -1.6534e-04,\n","         2.7430e-04, -3.4006e-04, -4.4932e-04, -7.9713e-04,  5.9697e-23,\n","         1.5321e-04, -1.4524e-03,  1.1002e-09, -3.9651e-04,  1.0817e-19,\n","         1.2528e-04, -5.5039e-04, -1.3199e-04,  1.3071e-04, -2.8005e-04,\n","        -2.0326e-07, -1.5963e-04, -1.0692e-03,  5.8280e-07,  1.6435e-04,\n","        -1.8342e-05, -3.8147e-04,  8.0908e-05, -1.0938e-03, -3.8426e-04,\n","         3.2438e-09, -9.8774e-04, -3.3271e-04,  4.3442e-07,  4.8637e-04,\n","        -1.3441e-03, -1.4611e-04,  6.6602e-04, -3.4148e-04, -3.5949e-16,\n","         1.4815e-04,  1.0952e-03,  2.7226e-04, -6.5695e-04,  9.1363e-05,\n","        -8.9821e-04,  4.4192e-04,  6.2850e-19,  3.8858e-04, -3.1687e-05,\n","         9.0202e-04, -5.2934e-04, -1.2301e-25,  4.7725e-06, -2.6684e-05,\n","         1.6954e-04,  2.3250e-04,  3.3384e-05,  4.2765e-04, -4.2044e-04,\n","        -3.0931e-04, -4.4658e-05,  5.1152e-04,  1.4371e-04, -1.1271e-03,\n","        -1.4662e-17,  1.5020e-04,  1.2763e-04, -4.5657e-25,  2.5384e-26,\n","         6.7952e-04, -2.8921e-04, -3.7298e-04,  8.9369e-05, -7.6164e-04,\n","        -4.5100e-22, -9.3968e-04,  5.7428e-05,  1.3570e-11, -1.1091e-04,\n","         6.0012e-08, -7.8395e-05, -1.5546e-04, -3.1933e-04, -5.1196e-04,\n","        -1.3355e-04, -8.2189e-04,  9.7401e-09, -7.4792e-04,  7.4326e-04,\n","         4.7402e-04, -1.6983e-23, -1.2083e-04, -2.7934e-04, -5.8190e-04,\n","        -1.0735e-03, -1.0669e-03, -2.4299e-05,  1.1058e-03,  1.5559e-03,\n","         8.4733e-04,  6.4316e-04, -2.0053e-21, -5.1105e-10, -3.7129e-04,\n","         4.7050e-16,  1.5679e-16, -5.3784e-04, -1.2932e-25,  7.7692e-05,\n","         2.0794e-04,  3.9820e-04, -7.0092e-04,  8.5799e-05, -1.2462e-29,\n","        -6.9857e-05, -9.3563e-22,  3.0685e-04,  7.1950e-08, -4.8236e-04,\n","        -3.9349e-04, -4.4700e-04, -2.2209e-04,  4.4463e-04, -3.8336e-04,\n","        -7.4164e-04, -5.3587e-04, -8.2744e-04,  4.1812e-04, -4.3667e-05,\n","        -4.1170e-04, -3.8319e-04, -1.8002e-04,  6.2637e-05,  2.9397e-23,\n","         7.8886e-04, -9.4106e-04, -9.8512e-05,  9.6690e-05,  1.4786e-21,\n","         4.2033e-04,  5.0294e-05,  3.9183e-07,  1.8213e-04, -3.3620e-04,\n","         1.6865e-13, -7.4370e-06, -3.7600e-04, -6.4815e-04, -2.9500e-04,\n","         1.5168e-04, -3.0816e-04, -9.5483e-04,  1.2120e-03, -2.7947e-12,\n","        -5.5472e-04, -2.4120e-04, -4.9942e-04, -5.7464e-05, -2.7443e-04,\n","        -5.5492e-04, -6.1600e-04, -3.7506e-04, -3.1769e-04,  1.1040e-04,\n","         2.4763e-24,  2.9090e-05, -1.1274e-09,  4.5547e-05,  4.3924e-04,\n","        -4.9909e-04, -2.7924e-04, -1.4987e-04, -2.1827e-04, -5.1466e-04,\n","         2.6885e-04, -4.8819e-27,  1.1195e-18, -7.9171e-04,  2.1534e-10,\n","         2.4136e-06,  1.7230e-09, -6.0350e-04, -4.3532e-04, -6.5198e-04,\n","        -2.7086e-05,  3.4209e-24,  7.6070e-04, -7.2515e-04,  7.5596e-04,\n","         3.5141e-04,  4.7451e-04,  5.0183e-04,  3.8343e-04, -4.6524e-04,\n","        -3.0794e-04,  5.9356e-06, -2.7536e-04, -1.7631e-04, -6.1661e-04,\n","        -4.4070e-04, -1.4984e-05,  6.3618e-04, -3.3979e-04, -1.4972e-03,\n","         1.4538e-04,  7.7935e-04, -4.9867e-07, -8.2443e-04, -3.4482e-04,\n","        -9.5255e-04,  3.2997e-04,  1.7053e-13,  3.5584e-04, -1.3345e-04,\n","         2.6274e-17, -1.8298e-03, -8.0168e-22,  3.1565e-13, -7.4259e-04,\n","         3.4792e-04,  4.8507e-05,  1.1892e-04, -4.3064e-04, -4.8121e-04,\n","        -3.0809e-04, -1.6668e-04, -1.0727e-03,  6.3413e-04, -5.9455e-11,\n","        -1.9977e-04,  3.5795e-04, -7.0719e-05,  1.5652e-04, -7.1746e-05,\n","        -4.5965e-04,  9.7004e-04, -5.5775e-04, -6.8025e-04,  7.5038e-05,\n","         1.4014e-04, -2.4902e-10,  2.0704e-04, -3.1551e-09,  1.9936e-04,\n","         1.1222e-04,  1.3376e-17,  5.9658e-10, -2.9549e-07,  3.2442e-04],\n","       device='mps:0'), 'exp_avg_sq': tensor([2.4898e-06, 2.6673e-06, 2.9130e-06, 2.2058e-06, 1.7996e-06, 3.2426e-06,\n","        1.7344e-06, 2.4948e-06, 1.4586e-06, 2.4244e-06, 1.4446e-06, 2.6070e-06,\n","        2.3701e-06, 3.7162e-06, 2.4555e-06, 4.3906e-06, 3.7620e-06, 3.4316e-06,\n","        2.9437e-06, 3.4343e-06, 3.5769e-06, 3.0991e-06, 3.2889e-06, 3.9108e-06,\n","        3.0326e-06, 3.9771e-06, 1.3184e-06, 9.7446e-07, 1.0534e-06, 4.2111e-06,\n","        3.6454e-06, 3.0304e-06, 2.4821e-06, 2.3775e-06, 2.7840e-06, 3.4056e-06,\n","        1.8147e-06, 3.4079e-06, 1.0726e-06, 1.8413e-06, 2.1064e-06, 3.8415e-06,\n","        3.1357e-06, 4.0561e-06, 2.8457e-06, 2.7374e-06, 3.4156e-06, 2.6379e-06,\n","        1.1544e-06, 2.5346e-06, 2.7323e-06, 3.0942e-06, 3.6878e-06, 3.8017e-06,\n","        3.0356e-06, 2.6784e-06, 2.8916e-06, 1.9298e-06, 4.3792e-06, 1.6777e-06,\n","        3.0091e-06, 2.9413e-06, 1.0113e-06, 4.0318e-06, 2.3262e-06, 3.6736e-06,\n","        2.9051e-06, 1.9224e-06, 3.1874e-06, 1.2266e-06, 2.7010e-06, 2.9439e-06,\n","        2.5830e-06, 2.6431e-06, 2.4632e-06, 3.3399e-06, 4.3785e-06, 3.7835e-06,\n","        3.4211e-06, 3.5202e-06, 4.4195e-06, 3.4270e-06, 2.2040e-06, 1.9266e-06,\n","        1.3078e-06, 3.2073e-06, 1.7760e-06, 9.3088e-07, 2.6623e-06, 2.5836e-06,\n","        4.1380e-06, 2.5711e-06, 3.1917e-06, 3.6341e-06, 2.8395e-06, 3.2375e-06,\n","        3.8755e-06, 4.4391e-06, 3.1471e-06, 2.2442e-06, 1.9673e-06, 4.1700e-06,\n","        1.0899e-06, 2.9403e-06, 3.1465e-06, 2.3707e-06, 3.0553e-06, 3.4459e-06,\n","        2.8057e-06, 3.5077e-06, 3.2971e-06, 2.0465e-06, 2.8554e-06, 4.1025e-06,\n","        4.4395e-06, 4.0704e-06, 4.5740e-06, 3.0494e-06, 8.8428e-07, 3.4805e-06,\n","        2.9146e-06, 3.2793e-06, 3.7435e-06, 2.0548e-06, 2.1299e-06, 2.9953e-06,\n","        3.1794e-06, 4.0008e-06, 2.5782e-06, 4.6346e-06, 3.5473e-06, 2.9860e-06,\n","        3.3700e-06, 3.6117e-06, 3.5646e-06, 3.0178e-06, 2.5897e-06, 1.9087e-06,\n","        1.4768e-06, 2.7377e-06, 2.7119e-06, 3.3380e-06, 2.9611e-06, 4.3599e-06,\n","        2.9529e-06, 2.9983e-06, 1.5486e-06, 4.1959e-06, 9.1371e-07, 2.5318e-06,\n","        1.2961e-06, 3.1253e-06, 3.0438e-06, 3.4660e-06, 1.8318e-06, 2.2696e-06,\n","        2.2377e-06, 2.2910e-06, 2.4533e-06, 2.2963e-06, 1.8726e-06, 4.1411e-06,\n","        1.3232e-06, 2.4498e-06, 3.0145e-06, 3.6896e-06, 3.0900e-06, 3.6506e-06,\n","        2.9337e-06, 1.4869e-06, 3.3716e-06, 2.7968e-06, 3.9747e-06, 2.7984e-06,\n","        2.8520e-06, 4.3303e-06, 3.0790e-06, 2.5062e-06, 2.2122e-06, 2.5422e-06,\n","        3.2688e-06, 2.6665e-06, 2.9762e-06, 1.4424e-06, 2.7502e-06, 2.9877e-06,\n","        2.8972e-06, 3.7743e-06, 2.8912e-06, 4.0916e-06, 2.6502e-06, 2.6239e-06,\n","        3.3241e-06, 3.2593e-06, 4.6702e-06, 3.3041e-06, 2.8762e-06, 2.6438e-06,\n","        2.9544e-06, 4.1204e-06, 2.7860e-06, 2.3142e-06, 2.6854e-06, 4.1156e-06,\n","        1.0655e-06, 2.5302e-06, 3.2031e-06, 3.4932e-06, 2.8981e-06, 1.7217e-06,\n","        9.1214e-07, 2.2845e-06, 5.4345e-07, 4.3803e-06, 2.2921e-06, 1.7512e-06,\n","        4.1777e-06, 3.4690e-06, 1.8834e-06, 3.0157e-06, 2.9560e-06, 3.5770e-06,\n","        2.8103e-06, 2.5244e-06, 3.3807e-06, 3.1940e-06, 3.1524e-06, 2.6882e-06,\n","        2.0350e-06, 3.3589e-06, 2.2392e-06, 3.7927e-06, 3.5240e-06, 3.5254e-06,\n","        2.5553e-06, 2.0118e-06, 3.4561e-06, 1.8813e-06, 2.4196e-06, 3.0658e-06,\n","        2.5827e-06, 2.6832e-06, 1.3736e-06, 3.2227e-06, 3.2301e-06, 3.4977e-06,\n","        3.7067e-06, 3.2577e-06, 1.5975e-06, 1.4878e-06, 2.5369e-06, 2.4880e-06,\n","        2.6602e-06, 3.7786e-06, 3.0197e-06, 3.4517e-06, 2.5202e-06, 1.6604e-06,\n","        3.4692e-06, 1.8676e-06, 1.7827e-06, 2.9428e-06, 2.4781e-06, 2.0355e-06,\n","        2.3548e-06, 2.5966e-06, 3.6988e-06, 2.6598e-06, 3.7513e-06, 2.5077e-06,\n","        3.6092e-06, 1.9216e-06, 3.4586e-06, 2.4256e-06, 2.9192e-06, 2.9165e-06,\n","        2.5761e-06, 2.5985e-06, 2.4758e-06, 2.2463e-06, 3.1913e-06, 2.2912e-06,\n","        3.6655e-06, 3.5517e-06, 2.1327e-06, 2.1968e-06, 3.6618e-06, 2.0180e-06,\n","        2.9151e-06, 2.9844e-06, 2.9714e-06, 1.8100e-06, 2.8954e-06, 2.9502e-06,\n","        1.4935e-06, 3.1013e-06, 2.9230e-06, 2.5151e-06, 2.6867e-06, 1.2965e-06,\n","        2.9319e-06, 2.3733e-06, 1.9107e-06, 2.9470e-06, 1.1001e-06, 2.6855e-06,\n","        2.8490e-06, 2.8589e-06, 3.2566e-06, 1.6422e-06, 3.0468e-06, 1.2764e-06,\n","        1.4389e-06, 2.6369e-06, 3.5672e-06, 2.9878e-06, 2.5057e-06, 3.1134e-06,\n","        3.4112e-06, 3.2127e-06, 3.4548e-06, 2.6255e-06, 2.0191e-06, 2.5628e-06,\n","        3.2902e-06, 8.4233e-07, 2.4179e-06, 2.8328e-06, 4.3019e-06, 2.9056e-06,\n","        3.1962e-06, 4.2631e-06, 2.3476e-06, 3.4200e-06, 3.2302e-06, 3.3785e-06,\n","        4.2216e-06, 2.7344e-06, 3.2689e-06, 2.4629e-06, 1.0570e-06, 2.6949e-06,\n","        1.6663e-06, 3.8902e-06, 4.1495e-06, 2.8614e-06, 2.2579e-06, 5.2081e-06,\n","        2.6664e-06, 3.2463e-06, 3.2087e-06, 1.9916e-06, 3.6066e-06, 2.2693e-06,\n","        2.3526e-06, 3.9151e-06, 3.3341e-06, 1.5505e-06, 2.7178e-06, 3.2365e-06,\n","        2.0773e-06, 2.2702e-06, 2.2053e-06, 2.0439e-06, 1.2361e-06, 2.7889e-06,\n","        3.0241e-06, 2.7138e-06, 2.4478e-06, 2.5977e-06, 3.2189e-06, 2.8795e-06,\n","        1.3820e-06, 2.0687e-06, 3.3047e-06, 1.8810e-06, 1.4740e-06, 3.8134e-06,\n","        3.3536e-06, 2.6142e-06, 3.0941e-06, 3.2747e-06, 2.6752e-06, 3.6177e-06,\n","        2.0714e-06, 2.6397e-06, 8.7687e-07, 3.3429e-06, 2.6795e-06, 3.2449e-06,\n","        3.8691e-06, 3.0663e-06, 3.1581e-06, 1.8279e-06, 2.6365e-06, 3.0758e-06,\n","        3.8062e-06, 3.4918e-06, 3.2267e-06, 2.6021e-06, 3.3852e-06, 1.2877e-06,\n","        3.1670e-06, 2.5393e-06, 2.0067e-06, 3.5154e-06, 2.5783e-06, 3.4629e-06,\n","        2.6067e-06, 4.2779e-06, 2.4027e-06, 3.2762e-06, 3.7296e-06, 2.7859e-06,\n","        2.9390e-06, 2.6408e-06, 2.7051e-06, 2.1761e-06, 2.5816e-06, 4.0497e-06,\n","        4.8336e-06, 3.9891e-06, 4.6977e-06, 2.2429e-06, 2.2575e-06, 3.0352e-06,\n","        2.7194e-06, 3.3402e-06, 2.4124e-06, 1.6409e-06, 3.5094e-06, 2.7356e-06,\n","        3.4695e-06, 2.7741e-06, 2.3989e-06, 2.3821e-06, 4.0765e-06, 3.5491e-06,\n","        3.2779e-06, 2.7473e-06, 1.0864e-06, 3.1056e-06, 4.7886e-06, 2.1533e-06,\n","        3.1340e-06, 2.2043e-06, 2.8111e-06, 3.7265e-06, 2.5741e-06, 2.6821e-06,\n","        2.2112e-06, 2.1358e-06, 1.7775e-06, 1.4247e-06, 2.0633e-06, 1.3597e-06,\n","        3.8650e-06, 2.9262e-06, 2.5140e-06, 2.2600e-06, 3.4308e-06, 2.3971e-06,\n","        3.0209e-06, 2.3258e-06, 2.4387e-06, 2.6525e-06, 2.7308e-06, 3.3104e-06,\n","        3.1967e-06, 3.0381e-06, 3.3952e-06, 3.3971e-06, 1.5074e-06, 2.9346e-06,\n","        2.5772e-06, 2.8073e-06, 2.9480e-06, 2.2539e-06, 2.7371e-06, 3.0832e-06,\n","        2.5948e-06, 3.3335e-06, 3.8687e-06, 2.5386e-06, 2.4388e-06, 3.5949e-06,\n","        2.7306e-06, 1.6785e-06, 3.6037e-06, 2.7433e-06, 2.8228e-06, 3.4413e-06,\n","        2.7959e-06, 3.5096e-06, 1.2205e-06, 3.8871e-06, 3.2994e-06, 3.3542e-06,\n","        3.4240e-06, 3.2944e-06, 3.7029e-06, 1.9382e-06, 1.3352e-06, 2.0855e-06,\n","        1.3309e-06, 3.0298e-06, 2.8751e-06, 9.7283e-07, 2.9103e-06, 2.3935e-06,\n","        1.3835e-06, 9.5732e-07, 3.0945e-06, 3.2746e-06, 1.3262e-06, 2.5102e-06,\n","        3.3032e-06, 1.8384e-06, 2.7906e-06, 8.9570e-07, 2.5049e-06, 2.7453e-06,\n","        4.0206e-06, 2.5386e-06, 2.6739e-06, 3.4230e-06, 3.2367e-06, 2.6227e-06,\n","        3.1064e-06, 3.1222e-06, 1.6394e-06, 2.8771e-06, 2.8488e-06, 3.2585e-06,\n","        1.9301e-06, 2.7849e-06, 2.5972e-06, 2.4590e-06, 2.6926e-06, 3.7369e-06,\n","        3.3608e-06, 3.0346e-06, 3.2346e-06, 2.8414e-06, 3.2661e-06, 2.4468e-06,\n","        2.3855e-06, 4.0774e-06, 2.9540e-06, 3.1249e-06, 2.7407e-06, 3.7555e-06,\n","        4.1112e-06, 3.5508e-06, 3.0869e-06, 3.3248e-06, 2.4264e-06, 2.1254e-06,\n","        1.9957e-06, 3.3975e-06, 2.9658e-06, 3.0621e-06, 3.3524e-06, 2.6131e-06,\n","        1.5285e-06, 2.6327e-06, 1.5881e-06, 2.2405e-06, 2.6760e-06, 1.2270e-06,\n","        2.0631e-06, 3.0480e-06, 2.6909e-06, 1.6960e-06, 1.7465e-06, 1.1538e-06,\n","        3.3783e-06, 1.1586e-06, 3.3381e-06, 2.3287e-06, 2.3691e-06, 3.7985e-06,\n","        2.8017e-06, 5.2001e-06, 2.4861e-06, 4.1778e-06, 2.6284e-06, 3.1645e-06,\n","        3.5177e-06, 9.7903e-07, 2.9372e-06, 3.5306e-06, 1.9820e-06, 2.8397e-06,\n","        3.1997e-06, 4.5438e-06, 3.2838e-06, 3.3485e-06, 2.1835e-06, 3.4904e-06,\n","        3.7176e-06, 2.9617e-06, 2.1934e-06, 3.7249e-06, 2.8479e-06, 1.4509e-06,\n","        3.9577e-06, 2.5905e-06, 2.1706e-06, 2.9408e-06, 3.7373e-06, 2.5133e-06,\n","        1.3313e-06, 2.2525e-06, 3.6460e-06, 4.9689e-06, 3.5696e-06, 2.9840e-06,\n","        2.9544e-06, 2.9515e-06, 4.0560e-06, 3.3349e-06, 2.4097e-06, 3.1253e-06,\n","        2.4377e-06, 1.3874e-06, 4.6653e-06, 3.3628e-06, 3.7215e-06, 2.3287e-06,\n","        3.5999e-06, 3.8286e-06, 2.7218e-06, 7.5371e-07, 2.5605e-06, 3.5285e-06,\n","        3.7224e-06, 3.5123e-06, 1.4783e-06, 3.1771e-06, 6.7063e-07, 2.3164e-06,\n","        1.6788e-06, 3.3804e-06, 2.4797e-06, 3.3459e-06, 2.3837e-06, 2.3871e-06,\n","        3.6113e-06, 3.5989e-06, 3.1352e-06, 3.0744e-06, 1.5752e-06, 3.4861e-06,\n","        9.5340e-07, 2.8398e-06, 1.6509e-06, 3.4902e-06, 3.5076e-06, 3.0971e-06,\n","        2.5925e-06, 2.7395e-06, 2.7647e-06, 3.5069e-06, 2.4875e-06, 2.9802e-06,\n","        3.5100e-06, 1.1435e-06, 3.2993e-06, 3.6118e-06, 2.2251e-06, 1.4445e-06,\n","        2.8564e-06, 1.6577e-06, 1.8328e-06, 3.3560e-06, 2.3157e-06, 2.2116e-06,\n","        1.9886e-06, 3.5563e-06, 1.7087e-06, 4.5795e-06, 2.7586e-06, 3.2150e-06,\n","        2.4689e-06, 2.8835e-06, 4.6408e-06, 2.1902e-06, 3.5451e-06, 3.1271e-06,\n","        8.5093e-07, 3.3100e-06, 3.0524e-06, 1.6644e-06, 3.2919e-06, 3.2313e-06,\n","        4.2665e-06, 3.4267e-06, 3.7527e-06, 2.1409e-06, 2.2705e-06, 2.3321e-06,\n","        4.8953e-06, 1.6542e-06, 2.0008e-06, 2.9769e-06, 3.2729e-06, 2.7518e-06,\n","        2.1772e-06, 2.8722e-06, 2.7605e-06, 2.9465e-06, 3.8409e-06, 2.4350e-06,\n","        2.8362e-06, 1.9524e-06, 1.6946e-06, 2.5170e-06, 3.3751e-06, 3.3012e-06,\n","        2.8098e-06, 3.4198e-06, 2.5132e-06, 2.8415e-06, 2.8176e-06, 2.7365e-06,\n","        2.1787e-06, 1.4903e-06, 3.6246e-06, 3.1480e-06, 3.0071e-06, 2.0272e-06,\n","        3.6082e-06, 2.9030e-06, 2.6397e-06, 4.8328e-06, 2.7624e-06, 3.3849e-06,\n","        3.5082e-06, 3.2793e-06, 4.1027e-06, 2.8587e-06, 3.2038e-06, 2.2375e-06,\n","        3.1666e-06, 1.7068e-06, 2.9527e-06, 2.9593e-06, 2.4966e-06, 1.1524e-06,\n","        3.8037e-06, 3.2349e-06, 1.7866e-06, 2.1601e-06, 3.6790e-06, 2.8733e-06,\n","        3.6029e-06, 2.0782e-06, 4.0682e-06, 1.6007e-06, 3.5969e-06, 4.7855e-06,\n","        2.0207e-06, 3.1061e-06, 1.0188e-06, 3.3723e-06, 3.4833e-06, 3.4911e-06,\n","        2.6948e-06, 2.3801e-06, 2.7799e-06, 2.9449e-06, 3.3603e-06, 2.9519e-06,\n","        5.0273e-06, 2.7552e-06, 2.5053e-06, 2.5478e-06, 3.7153e-06, 3.1854e-06,\n","        1.8147e-06, 3.7277e-06, 2.9769e-06, 2.1262e-06, 2.7481e-06, 3.5108e-06,\n","        2.2448e-06, 2.6661e-06, 4.2841e-06, 9.3823e-07, 3.1694e-06, 3.5027e-06,\n","        4.6311e-06, 3.0537e-06, 2.3557e-06, 3.3143e-06, 3.8644e-06, 1.5462e-06,\n","        3.0686e-06, 2.9983e-06, 3.2887e-06, 2.2715e-06, 1.0353e-06, 1.9792e-06,\n","        3.4257e-06, 2.4920e-06, 3.0058e-06, 2.2348e-06, 2.4146e-06, 4.1034e-06,\n","        3.0165e-06, 3.0203e-06, 2.8008e-06, 2.0430e-06, 3.8679e-06, 1.8558e-06,\n","        3.4735e-06, 2.6611e-06, 1.4057e-06, 1.1596e-06, 2.6206e-06, 2.6640e-06,\n","        2.8362e-06, 3.8896e-06, 2.5080e-06, 8.4435e-07, 3.4511e-06, 2.1614e-06,\n","        1.6002e-06, 3.4542e-06, 2.3469e-06, 2.9084e-06, 2.9763e-06, 2.4699e-06,\n","        3.8231e-06, 2.7279e-06, 2.5349e-06, 2.0411e-06, 2.8829e-06, 2.6747e-06,\n","        2.3084e-06, 1.3182e-06, 3.5064e-06, 3.1066e-06, 3.2704e-06, 3.7499e-06,\n","        3.2240e-06, 1.5594e-06, 3.6295e-06, 3.7961e-06, 3.9524e-06, 2.6776e-06,\n","        1.4186e-06, 1.7008e-06, 2.2244e-06, 1.1235e-06, 1.2213e-06, 2.8540e-06,\n","        1.3854e-06, 2.6086e-06, 2.7229e-06, 3.4843e-06, 3.6924e-06, 3.2953e-06,\n","        1.0323e-06, 3.4913e-06, 1.6451e-06, 2.5825e-06, 2.9326e-06, 5.5273e-06,\n","        2.9579e-06, 4.1380e-06, 3.1001e-06, 2.9085e-06, 3.6574e-06, 2.8902e-06,\n","        3.3084e-06, 2.5737e-06, 3.2249e-06, 1.8353e-06, 3.6206e-06, 2.0098e-06,\n","        3.0390e-06, 4.7386e-06, 9.3860e-07, 2.8908e-06, 4.1973e-06, 2.8834e-06,\n","        3.5647e-06, 1.6357e-06, 3.5615e-06, 3.0982e-06, 2.2657e-06, 2.6355e-06,\n","        2.5180e-06, 2.0310e-06, 3.1115e-06, 2.6094e-06, 2.3333e-06, 2.1509e-06,\n","        4.1874e-06, 2.4347e-06, 2.9491e-06, 3.4024e-06, 1.9377e-06, 3.8139e-06,\n","        4.3632e-06, 3.3946e-06, 2.7024e-06, 3.0508e-06, 4.0543e-06, 3.8227e-06,\n","        2.7630e-06, 3.9522e-06, 3.6226e-06, 1.2073e-06, 2.1558e-06, 3.1393e-06,\n","        3.0263e-06, 3.3496e-06, 3.1873e-06, 2.6785e-06, 2.9450e-06, 3.0916e-06,\n","        3.6019e-06, 2.9647e-06, 7.6273e-07, 1.6957e-06, 3.4713e-06, 2.1016e-06,\n","        2.9022e-06, 2.5914e-06, 2.1720e-06, 2.2983e-06, 3.4299e-06, 3.2182e-06,\n","        8.3787e-07, 3.1857e-06, 3.4009e-06, 3.1433e-06, 2.5281e-06, 2.8513e-06,\n","        4.4084e-06, 3.3390e-06, 3.7369e-06, 3.2319e-06, 1.3574e-06, 2.9634e-06,\n","        3.4201e-06, 3.4910e-06, 3.7223e-06, 3.3280e-06, 2.7897e-06, 3.9584e-06,\n","        3.2512e-06, 3.2055e-06, 2.0695e-06, 2.1449e-06, 3.0343e-06, 2.6185e-06,\n","        2.5075e-06, 2.8388e-06, 2.6574e-06, 2.4454e-06, 3.8948e-06, 1.9824e-06,\n","        3.0055e-06, 9.1277e-07, 3.0174e-06, 2.8647e-06, 3.4795e-06, 2.9618e-06,\n","        3.4329e-06, 2.8557e-06, 3.7836e-06, 4.2320e-06, 3.2527e-06, 3.7895e-06,\n","        2.8168e-06, 1.8700e-06, 2.7000e-06, 3.8626e-06, 2.2452e-06, 4.8642e-06,\n","        1.8969e-06, 2.6543e-06, 3.9925e-06, 2.4259e-06, 3.1667e-06, 3.0861e-06,\n","        3.4649e-06, 2.5722e-06, 3.3259e-06, 2.2073e-06, 2.5093e-06, 2.7568e-06,\n","        1.3682e-06, 2.6153e-06, 2.8476e-06, 2.2370e-06], device='mps:0')}, 5: {'step': tensor(1025.), 'exp_avg': tensor([ 1.4987e-04, -2.7279e-04,  1.0473e-03,  1.7228e-04, -5.9393e-04,\n","         3.1653e-04, -4.8767e-04, -3.5994e-04, -1.4423e-04,  4.3694e-04,\n","        -3.7577e-05, -1.0390e-03, -2.5919e-04,  4.3304e-04, -2.5175e-04,\n","         7.5169e-04, -1.4907e-03,  1.1379e-03,  5.1851e-04,  1.3048e-04,\n","        -1.8456e-04,  8.2323e-04,  1.3741e-04, -6.1856e-04, -7.7116e-04,\n","        -4.6140e-04,  3.2480e-05,  8.7329e-04, -8.5942e-04,  3.9512e-04,\n","        -3.1190e-04, -2.5513e-04, -1.2568e-04,  7.0731e-04, -8.7685e-04,\n","         1.2803e-04, -1.6150e-04,  4.5139e-04, -1.5677e-04,  1.0826e-04,\n","        -4.3418e-05,  1.1782e-03, -5.5816e-05, -4.1562e-04, -1.2816e-03,\n","        -1.3550e-03,  6.5365e-04, -8.7162e-05, -2.0152e-04,  5.2756e-04,\n","        -5.9966e-05, -8.4597e-04,  9.0732e-04, -2.1434e-04,  4.0486e-04,\n","        -1.7316e-04, -3.9141e-04,  5.1378e-04,  8.4660e-04,  5.2730e-04,\n","         1.3821e-04,  1.5926e-04,  7.3177e-05,  4.2852e-04, -1.3104e-03,\n","         4.0021e-04,  4.8681e-04,  1.8702e-04, -1.0265e-04, -8.4737e-04,\n","         2.6720e-04, -5.9159e-04, -7.7967e-04, -2.8364e-04, -7.3690e-04,\n","         2.4407e-04, -3.6478e-04, -9.7330e-05,  2.0759e-04, -8.2360e-04,\n","        -3.8376e-04, -3.9379e-04,  1.8955e-04, -9.5886e-04,  8.8058e-04,\n","         5.4031e-05,  4.9628e-04,  9.2596e-04,  4.5954e-04,  4.7291e-04,\n","        -3.4329e-04,  8.6441e-04, -1.0505e-03,  5.5317e-04, -1.6858e-04,\n","        -3.9731e-04,  7.6613e-04, -4.5471e-04, -6.8855e-04, -3.3126e-04,\n","        -1.2009e-05,  2.6215e-04,  1.2647e-03,  3.0321e-04, -3.9819e-04,\n","         2.1506e-05, -5.2364e-05,  4.4293e-04, -1.2700e-03,  5.0086e-04,\n","         7.0136e-04, -4.7215e-04,  1.0718e-03, -6.0525e-04, -4.2773e-04,\n","        -6.7591e-04,  3.3655e-04,  5.0820e-04, -1.1162e-03,  1.9918e-04,\n","         5.4009e-04,  6.8767e-04, -3.9641e-04,  9.9990e-04,  1.7294e-04,\n","         1.9630e-04, -1.4111e-04,  3.6817e-04, -5.7639e-05, -1.9971e-04,\n","         1.4944e-03, -7.9235e-04,  1.4844e-04, -4.5372e-04, -3.3827e-04,\n","        -4.3062e-05,  5.8944e-04, -8.4387e-04, -2.4907e-04,  1.1479e-03,\n","        -2.1842e-04,  2.4209e-04,  2.9574e-04, -3.2937e-04,  1.1407e-03,\n","        -1.6297e-04,  2.0158e-04, -8.2648e-04,  1.2334e-04, -7.4500e-04,\n","         3.5390e-04, -2.9769e-04, -6.7713e-04, -1.8736e-04, -6.9215e-04,\n","        -2.3635e-05, -8.1663e-04, -1.0536e-03, -1.1051e-03,  7.1615e-05,\n","        -4.5549e-04, -1.1108e-03, -9.7335e-05,  1.2684e-03, -2.2716e-05,\n","        -2.5549e-04, -5.4419e-04,  8.8164e-05,  5.5970e-04, -3.4180e-05,\n","         8.2215e-04,  4.9188e-04,  2.2803e-05,  8.7019e-04, -6.9001e-04,\n","        -3.0058e-04,  4.2325e-04, -4.2057e-04,  2.5545e-05,  3.6820e-04,\n","        -5.6693e-04, -2.0903e-04, -2.1457e-04, -3.4714e-04, -3.9576e-04,\n","         5.2959e-04,  1.3265e-03,  4.4243e-04, -6.3350e-04,  5.3484e-04,\n","         1.1881e-03,  2.0601e-04, -8.6117e-05,  3.0616e-04, -1.0596e-04,\n","         1.2705e-03, -2.0387e-04, -2.8705e-04,  8.6573e-04,  2.7032e-04,\n","        -4.7173e-05,  9.9805e-05,  8.7570e-04, -8.9513e-04, -3.0286e-04,\n","        -6.9320e-04,  2.8021e-04,  5.4862e-04, -3.0560e-04, -1.4026e-04,\n","        -2.4816e-05,  7.2448e-04,  2.8665e-04, -8.5876e-04, -9.9456e-04,\n","         1.0459e-03,  3.9993e-04,  4.9991e-04,  6.3831e-04,  5.7771e-04,\n","        -6.6355e-04, -2.1108e-04,  9.6541e-04, -2.0390e-04, -5.1747e-04,\n","        -4.7010e-05, -4.1336e-05, -2.4752e-04, -1.5379e-04, -2.4503e-04,\n","        -4.9348e-04, -1.0037e-03, -6.2763e-05,  3.0956e-04,  4.6117e-04,\n","         2.8296e-04,  1.4835e-05,  2.5678e-04, -2.1780e-04,  2.0527e-04,\n","         1.5321e-04,  5.5045e-04, -3.1125e-05,  7.5062e-04, -4.1861e-04,\n","         1.6157e-04, -7.4835e-04,  8.6868e-05,  9.8239e-04,  1.3487e-04,\n","        -4.3013e-04, -4.0296e-04,  5.1147e-04, -1.0752e-03,  5.1412e-04,\n","         1.0677e-03, -2.2135e-04,  3.6969e-04, -7.7135e-04,  6.0764e-05,\n","         6.1505e-06,  5.9194e-04,  5.8940e-05, -1.7997e-04,  7.4849e-04,\n","         2.4090e-04,  1.4258e-03,  3.5956e-04,  7.4323e-04, -7.8852e-04,\n","         7.9214e-04,  3.1366e-05,  2.4760e-04,  7.7056e-04, -7.3306e-05,\n","         3.4519e-04,  1.0189e-03,  1.9040e-04,  8.2644e-04,  2.3250e-04,\n","        -7.2800e-04, -3.6543e-04, -1.7819e-04, -8.0167e-04, -4.3989e-04,\n","         8.7440e-04, -1.6520e-03,  5.2579e-04, -3.6872e-04,  1.4162e-04,\n","        -1.1486e-04, -1.0432e-03,  3.6062e-04,  9.5730e-05,  5.7527e-04,\n","         8.0440e-04,  5.4418e-04,  2.2077e-04,  6.9799e-04,  1.2059e-03,\n","        -7.7864e-04,  1.4997e-03, -7.9351e-04, -1.4560e-04,  3.1885e-04,\n","         7.3596e-06,  2.1880e-06,  2.9370e-04, -2.8721e-04, -1.0356e-03,\n","        -2.3049e-04, -1.0192e-03, -9.0655e-05,  7.1823e-05,  4.2705e-04,\n","         3.5771e-04,  8.4155e-05, -3.8306e-04,  4.3931e-05,  3.9227e-04,\n","         5.1401e-04, -3.3390e-04, -2.1646e-04,  5.6846e-05, -1.0461e-03,\n","         1.1513e-03, -4.5242e-04, -5.3182e-04,  1.9762e-04, -7.7083e-04,\n","         3.5747e-04, -5.0844e-05,  5.6491e-04,  1.2570e-04, -7.1315e-04,\n","        -9.5041e-04,  1.2474e-03,  4.0524e-04, -2.2727e-04,  2.6282e-04,\n","         3.1596e-04,  8.2712e-05, -2.3013e-04,  3.8284e-04, -1.2562e-04,\n","        -6.7924e-04, -7.2955e-05, -3.6861e-04,  3.9279e-05,  1.0863e-03,\n","         1.2751e-04,  2.5368e-05, -1.8229e-04,  9.0344e-05, -1.8924e-04,\n","         1.7942e-04,  7.1681e-05,  9.0801e-04,  1.2856e-04, -2.3965e-04,\n","        -6.8646e-04, -3.1152e-05, -8.3008e-05, -2.7838e-04, -4.7732e-04,\n","         6.8974e-04,  4.2959e-04,  5.0643e-04, -7.2514e-05, -4.0792e-05,\n","         8.2374e-04, -1.0915e-03, -3.7432e-04,  6.4387e-04,  4.8345e-04,\n","         1.9337e-04,  3.5130e-04,  9.7885e-07, -3.8502e-04, -6.4965e-04,\n","         5.8476e-04, -3.8552e-04, -1.6432e-03,  9.5780e-04,  3.1802e-04,\n","        -2.3720e-05,  5.6457e-05,  4.3195e-04,  8.5266e-04, -2.2628e-04,\n","        -6.2723e-04, -4.1369e-04, -1.7160e-04,  2.0196e-04, -2.9059e-04,\n","         4.0966e-04,  3.0884e-04, -9.3769e-06,  3.8603e-05, -6.3837e-04,\n","        -5.0044e-04,  5.9467e-04, -1.3137e-03,  5.8269e-05, -3.1298e-04,\n","         6.2267e-04,  6.1461e-04, -9.2269e-04, -5.3351e-04,  1.6554e-04,\n","         8.9497e-04, -7.5249e-04,  8.9336e-05, -3.3879e-05,  4.2935e-04,\n","        -8.6048e-04,  6.1803e-04,  6.3614e-04, -1.7852e-04, -6.1104e-04,\n","        -2.2073e-04, -7.5069e-06,  1.7598e-04,  1.9798e-04, -1.8167e-04,\n","        -6.1991e-04,  3.2199e-04,  7.0944e-04, -1.3099e-03,  2.9802e-04,\n","        -4.2123e-04, -4.0839e-04, -4.2405e-04, -1.5722e-05,  5.6240e-06,\n","         9.6897e-04,  4.6350e-04, -6.3788e-04, -9.9662e-04, -1.5336e-03,\n","        -7.2748e-04, -9.1821e-04, -2.5044e-04,  7.4525e-04,  9.2661e-04,\n","         1.1804e-05, -3.2577e-04, -1.2726e-03, -1.3253e-03, -5.1339e-04,\n","         6.8580e-04, -7.0071e-04, -1.1183e-03,  6.0517e-05, -1.4317e-04,\n","         1.7107e-04,  7.2258e-04, -3.2567e-04,  3.9596e-04, -4.8293e-04,\n","         9.3906e-04,  7.8664e-06, -1.0226e-03,  3.6817e-04, -9.1922e-04,\n","        -1.8008e-04, -4.9367e-04,  1.4220e-05, -3.4377e-04,  1.3433e-04,\n","         2.9401e-04,  5.3094e-04,  7.9937e-04,  4.9104e-04,  7.6661e-04,\n","         3.0714e-04, -4.2507e-04,  8.9302e-06, -1.4161e-04, -9.7224e-04,\n","        -5.6203e-05,  1.6770e-04, -3.1371e-04,  3.1216e-05,  8.0443e-04,\n","        -2.5933e-04,  6.9197e-04, -2.7416e-04, -1.6894e-04,  1.0679e-03,\n","         4.6228e-04, -1.4401e-04, -1.0076e-04,  4.8430e-04,  1.0536e-03,\n","         2.2232e-04,  4.9468e-04,  2.3570e-04, -1.1595e-04, -2.6442e-04,\n","         5.9386e-04,  2.6658e-04, -2.2727e-04,  5.9077e-04,  3.8194e-04,\n","        -3.0219e-04, -9.0617e-04,  2.8337e-04,  6.9085e-04,  1.3546e-03,\n","         8.6146e-04, -2.4284e-04, -3.9270e-04, -2.5297e-04,  6.6240e-05,\n","         7.1265e-04,  6.9466e-04,  3.1307e-04,  8.6419e-04,  8.4514e-04,\n","         7.4794e-04,  1.3900e-04, -3.2359e-04, -5.9133e-05, -2.8723e-04,\n","        -3.4160e-04,  9.9271e-04,  7.0005e-04, -8.6926e-05, -1.1840e-04,\n","        -7.1076e-04, -3.6860e-04, -5.7639e-04,  1.6038e-03,  7.8919e-05,\n","         9.9153e-04,  6.6739e-06, -1.5140e-06, -6.2240e-04, -3.7235e-04,\n","        -5.6460e-04, -7.2552e-04, -6.9133e-05,  6.1572e-04,  9.5568e-04,\n","         2.1398e-04, -3.8926e-04,  2.1957e-04, -1.5257e-04, -8.0992e-04,\n","         6.5396e-04,  1.1547e-03, -7.8357e-04, -8.0617e-04, -1.0542e-04,\n","         6.0065e-04,  1.1535e-04, -7.8683e-04,  4.4859e-05,  3.3832e-04,\n","        -2.2718e-05, -2.8435e-04, -9.0087e-04,  5.6576e-05, -2.5608e-04,\n","         2.0218e-04, -2.2734e-04, -1.8290e-04, -1.6473e-04, -1.2615e-04,\n","        -5.7793e-04, -7.4066e-04, -3.9227e-04, -5.4511e-04,  9.0472e-06,\n","         4.7629e-04, -5.6428e-04, -3.7940e-04, -1.2043e-03, -1.0260e-04,\n","        -1.8219e-04, -3.3870e-04, -4.3708e-04, -1.8885e-04,  2.4774e-04,\n","         7.6441e-04,  1.6247e-04, -1.0521e-03, -5.3271e-04,  7.4422e-04,\n","         9.6025e-05, -3.9447e-04, -3.8072e-05,  1.8096e-03,  1.6331e-04,\n","         5.4094e-04,  1.1739e-03, -2.9081e-04,  3.0217e-04,  2.9191e-04,\n","        -5.0609e-04,  5.2172e-04,  2.7793e-04,  1.9565e-04, -2.5954e-04,\n","        -8.7031e-04, -1.2666e-03, -2.4398e-04,  1.5882e-04,  8.0635e-04,\n","         4.7076e-04,  4.7339e-04,  3.4509e-04, -6.1094e-05, -6.1505e-04,\n","         8.9341e-06, -6.3074e-04, -3.6945e-04,  1.0666e-05,  4.1888e-04,\n","        -3.5212e-04,  2.2964e-04,  1.2416e-04,  2.1904e-04,  9.5917e-04,\n","        -4.1004e-04, -3.3563e-04,  3.9639e-04,  3.6059e-04,  3.6315e-04,\n","         3.0589e-04, -1.0217e-03,  6.7915e-04, -3.4613e-04,  5.6547e-04,\n","         5.8486e-04, -1.2297e-03,  4.4132e-04,  5.2215e-04, -2.4752e-04,\n","        -2.0751e-04,  7.0772e-04,  9.9779e-04, -1.2020e-03, -6.5384e-04,\n","        -1.3348e-03, -4.2307e-04, -5.8660e-04, -2.0248e-04,  1.1952e-04,\n","        -5.2204e-04,  3.3910e-04,  1.4404e-03, -1.2659e-04,  2.2206e-05,\n","         3.2888e-04,  6.9299e-04,  1.1054e-04,  2.6816e-04, -6.3908e-04,\n","         1.2712e-04,  8.7280e-04,  1.3308e-06, -3.9930e-05, -7.1934e-04,\n","         1.5612e-04,  3.2756e-04, -5.9888e-05,  7.3998e-04,  3.5829e-04,\n","         8.6536e-04, -1.1324e-03, -1.9660e-05, -2.5895e-05,  1.3339e-05,\n","         4.8987e-04, -9.7797e-04, -4.6733e-05, -1.3619e-04, -8.5135e-04,\n","         2.8169e-04, -6.7648e-04,  1.0408e-03, -1.2923e-03,  6.8828e-04,\n","         6.8363e-04,  5.4887e-04, -4.2072e-05,  4.4586e-04,  2.7836e-04,\n","         7.1911e-05, -1.4237e-04, -3.5631e-04,  1.6339e-04,  7.6068e-04,\n","        -2.8818e-04,  4.3285e-04, -1.8860e-04, -1.4817e-04, -1.1269e-03,\n","         7.7848e-04,  5.8588e-04,  2.9823e-05,  2.2688e-04, -2.3179e-04,\n","        -7.3711e-06, -2.6191e-04,  6.7857e-04, -6.2041e-04, -8.2966e-04,\n","        -2.0608e-04,  1.6710e-04, -7.1484e-05, -1.0918e-04,  4.5923e-04,\n","         1.1330e-03,  1.9822e-04, -1.1164e-03,  8.5716e-04, -1.1537e-04,\n","        -1.4143e-04, -1.2828e-04, -2.2677e-04, -5.0437e-04,  5.6471e-04,\n","         9.6473e-04, -6.7795e-05, -2.2288e-04,  2.9454e-04,  7.5913e-04,\n","        -5.0400e-04, -1.7731e-04, -2.6680e-04, -8.1225e-04, -7.5214e-05,\n","        -7.0824e-05,  1.6060e-04, -1.5832e-03,  9.6028e-05,  4.9866e-04,\n","        -1.5596e-04,  1.5511e-04, -3.0874e-05, -1.6060e-04,  4.8403e-04,\n","        -5.0780e-04,  2.3025e-04, -1.5761e-04, -2.5674e-04, -2.6058e-05,\n","         1.2442e-04,  6.3993e-04, -8.6761e-04,  5.0420e-04,  2.1396e-05,\n","         4.2517e-05, -5.5226e-04, -5.2458e-05, -5.8150e-04, -4.9587e-04,\n","         5.3166e-05, -7.3425e-04,  2.9265e-04, -2.3049e-04,  6.1536e-04,\n","        -4.0300e-04,  5.5359e-04, -8.7197e-04,  3.9326e-04,  6.8658e-05,\n","         1.3800e-04, -5.3887e-04,  1.0719e-04,  1.1734e-04, -4.7118e-05,\n","        -7.6816e-04, -6.3553e-05,  7.9050e-04, -1.2533e-03,  3.4941e-04,\n","         7.0763e-04, -1.7584e-04,  4.8733e-04, -9.0888e-04, -8.2169e-04,\n","         1.0921e-03,  2.5250e-04, -6.5880e-04, -3.5123e-04, -5.6397e-05,\n","        -3.2499e-04,  3.4478e-04,  5.8220e-04,  4.9311e-04,  5.3701e-04,\n","        -5.2157e-04,  4.9348e-04,  9.8569e-05, -3.6344e-04,  5.9417e-04,\n","         1.3272e-04, -2.1491e-04, -2.4918e-04, -5.8191e-04,  5.9952e-04,\n","        -6.6123e-04,  9.0224e-05,  1.5288e-03,  1.3586e-03, -6.8945e-04,\n","        -6.1677e-04,  4.6512e-04, -5.0347e-04, -4.5102e-04, -8.9110e-04,\n","        -2.3297e-04,  1.3963e-04, -3.5706e-04, -1.1383e-05,  8.1634e-04,\n","         8.2832e-04, -1.9862e-04,  7.0211e-04, -1.3579e-03,  5.7758e-05,\n","         2.8026e-05,  9.3397e-04, -4.8265e-05, -4.5945e-04,  2.2135e-04,\n","         3.4172e-04, -4.8879e-04,  2.3050e-04, -1.0285e-03, -4.7147e-04,\n","         2.8288e-04,  2.5713e-04,  4.7120e-04,  7.9075e-04,  3.1292e-04,\n","         6.7948e-04,  3.9855e-04,  5.0585e-04,  1.4869e-04, -2.5336e-04,\n","        -6.8562e-04, -1.6362e-04, -7.3544e-04,  1.4080e-03,  1.7157e-05,\n","        -3.8398e-04,  1.5713e-03, -4.7004e-04,  5.6148e-04,  1.1662e-03,\n","         3.6607e-04,  4.5220e-04, -8.4356e-04, -2.3656e-04,  8.9315e-05,\n","         2.1036e-04, -4.3467e-04, -2.9215e-04,  7.6310e-05,  5.8113e-04,\n","        -3.0132e-04,  6.1883e-04, -1.1047e-04, -5.1311e-04,  6.5611e-04,\n","         6.8071e-04,  4.6945e-04, -4.0614e-04, -8.3100e-05,  6.5194e-05,\n","        -9.8489e-04, -5.0963e-04,  4.3857e-05, -4.1766e-04,  2.5124e-04,\n","         1.7826e-04, -1.1199e-04,  5.0052e-04, -7.9889e-04,  6.9377e-04,\n","         1.1595e-04,  4.2316e-04,  7.0744e-04,  4.2742e-04, -3.2823e-04,\n","         9.0353e-04,  3.5750e-04, -2.3952e-04, -1.1474e-03, -6.4534e-04,\n","         5.2191e-04,  1.1043e-03, -6.5850e-05,  3.1148e-04,  5.5964e-04,\n","         8.1012e-05,  3.3398e-04,  6.2702e-04,  5.2419e-04,  9.1762e-04,\n","        -8.9316e-04, -4.8584e-04, -2.2845e-04, -3.7365e-04,  5.4052e-04,\n","         1.1119e-03,  3.8229e-04, -6.6571e-04, -7.7840e-04,  7.3938e-04,\n","        -5.3352e-04,  6.2336e-04,  1.2800e-03,  1.0668e-05,  8.2751e-04,\n","         3.9691e-04, -6.6931e-04, -1.3859e-04,  9.3425e-04, -7.2434e-04,\n","         3.7234e-05, -1.2891e-04,  1.0649e-04,  1.4523e-04, -1.6826e-05,\n","        -1.8305e-04,  3.6011e-06,  6.8793e-05,  3.6363e-04,  2.0950e-04,\n","         8.0993e-05,  2.1419e-04,  2.3977e-04,  1.0154e-03,  2.7391e-04,\n","        -6.4812e-04,  3.4125e-04, -2.2176e-04, -1.5886e-04,  3.7085e-04,\n","         8.1838e-04,  2.5516e-04,  6.7920e-04, -1.1366e-03, -5.5534e-04,\n","         4.4203e-04,  3.1547e-04,  5.4471e-04,  9.2414e-06, -4.4013e-04,\n","         1.1695e-03,  2.2822e-04,  1.5124e-04, -3.6772e-05,  1.2803e-04,\n","         3.0779e-04, -1.1732e-03, -9.3275e-04, -2.5178e-04, -9.5329e-04,\n","         5.9588e-04,  4.9357e-04, -1.3235e-04,  8.2705e-05, -2.9654e-04,\n","         2.7240e-04,  1.0089e-03, -8.4813e-04, -4.9126e-04,  3.7628e-04,\n","         4.6369e-04, -1.2231e-04,  1.1230e-03,  1.1764e-03,  4.1239e-04,\n","        -4.2425e-04, -5.0576e-04,  4.0480e-04, -2.3645e-04,  4.1607e-04,\n","         6.2600e-04, -9.7002e-04,  8.6312e-04, -1.0784e-03,  6.9948e-04,\n","        -6.8484e-04, -3.2632e-04,  1.9554e-04, -4.2369e-04,  7.2295e-04,\n","        -1.0991e-03, -1.0076e-03,  1.2084e-03,  1.5525e-04,  6.4814e-04],\n","       device='mps:0'), 'exp_avg_sq': tensor([2.5577e-06, 4.1537e-06, 3.4313e-06, 2.5459e-06, 4.1306e-06, 3.9348e-06,\n","        2.9555e-06, 3.3278e-06, 3.0679e-06, 3.4219e-06, 3.0044e-06, 3.1510e-06,\n","        3.3552e-06, 3.6780e-06, 3.5075e-06, 3.2812e-06, 3.9146e-06, 3.9329e-06,\n","        3.9306e-06, 3.5346e-06, 4.0572e-06, 3.5292e-06, 3.4259e-06, 4.1279e-06,\n","        3.6975e-06, 4.1810e-06, 3.0663e-06, 3.0018e-06, 2.7742e-06, 3.8167e-06,\n","        4.2490e-06, 3.2629e-06, 3.0857e-06, 2.8424e-06, 2.9702e-06, 3.6844e-06,\n","        3.0621e-06, 4.0522e-06, 3.6349e-06, 3.4849e-06, 3.3138e-06, 3.1941e-06,\n","        3.6974e-06, 3.9053e-06, 3.0761e-06, 3.4544e-06, 3.4489e-06, 3.2260e-06,\n","        3.3222e-06, 3.4309e-06, 3.3154e-06, 2.9660e-06, 3.8844e-06, 4.2019e-06,\n","        3.2938e-06, 3.2990e-06, 3.8068e-06, 3.4144e-06, 3.8599e-06, 2.9197e-06,\n","        3.7920e-06, 3.3865e-06, 2.6757e-06, 2.9726e-06, 3.7600e-06, 4.1067e-06,\n","        3.9011e-06, 3.6999e-06, 3.5209e-06, 3.3889e-06, 3.2486e-06, 2.7837e-06,\n","        4.3539e-06, 3.0366e-06, 4.0064e-06, 4.6600e-06, 3.9946e-06, 3.8014e-06,\n","        4.2565e-06, 3.9551e-06, 3.8051e-06, 3.9414e-06, 3.3655e-06, 3.1448e-06,\n","        2.7717e-06, 4.0037e-06, 3.5549e-06, 2.6626e-06, 2.9717e-06, 3.1843e-06,\n","        4.5224e-06, 3.2022e-06, 4.3826e-06, 4.2172e-06, 3.3896e-06, 4.2850e-06,\n","        4.7065e-06, 4.1088e-06, 3.9171e-06, 3.3868e-06, 2.9875e-06, 5.0161e-06,\n","        3.6383e-06, 2.9750e-06, 3.3444e-06, 2.8025e-06, 3.4733e-06, 3.6334e-06,\n","        3.7641e-06, 3.8457e-06, 3.8148e-06, 3.4877e-06, 3.7500e-06, 3.9925e-06,\n","        3.3890e-06, 3.5428e-06, 4.5143e-06, 3.7151e-06, 3.5091e-06, 3.3923e-06,\n","        3.7002e-06, 3.4904e-06, 4.0806e-06, 3.4313e-06, 3.5965e-06, 3.9545e-06,\n","        3.5315e-06, 4.6041e-06, 3.0113e-06, 4.3201e-06, 3.6634e-06, 2.8590e-06,\n","        3.9841e-06, 3.2697e-06, 3.5055e-06, 3.6055e-06, 2.9699e-06, 3.7035e-06,\n","        3.0475e-06, 3.7695e-06, 3.1287e-06, 3.7213e-06, 3.4242e-06, 3.6462e-06,\n","        3.5396e-06, 3.4074e-06, 2.7335e-06, 3.6130e-06, 3.3298e-06, 2.9383e-06,\n","        3.0234e-06, 3.0100e-06, 3.6196e-06, 4.3068e-06, 3.5557e-06, 2.5084e-06,\n","        2.8547e-06, 3.0157e-06, 4.0271e-06, 2.4930e-06, 3.2814e-06, 3.7483e-06,\n","        3.0524e-06, 2.9294e-06, 3.4439e-06, 3.6468e-06, 3.3773e-06, 2.9869e-06,\n","        3.0127e-06, 3.0502e-06, 3.2198e-06, 3.1146e-06, 4.7062e-06, 3.4479e-06,\n","        3.3887e-06, 4.3156e-06, 3.9320e-06, 4.0396e-06, 2.7125e-06, 3.2987e-06,\n","        3.7184e-06, 3.4561e-06, 2.8340e-06, 3.1422e-06, 3.1241e-06, 3.8229e-06,\n","        3.7605e-06, 3.4022e-06, 4.0860e-06, 4.4621e-06, 3.5741e-06, 3.2240e-06,\n","        3.4695e-06, 3.7040e-06, 4.4153e-06, 4.1114e-06, 3.4440e-06, 3.3232e-06,\n","        3.4394e-06, 3.7211e-06, 3.9829e-06, 3.3431e-06, 3.2653e-06, 3.4810e-06,\n","        3.0559e-06, 3.4402e-06, 3.5184e-06, 3.2417e-06, 3.3841e-06, 3.3452e-06,\n","        3.3964e-06, 3.9888e-06, 2.2668e-06, 3.4807e-06, 3.2913e-06, 3.2226e-06,\n","        3.5175e-06, 3.6285e-06, 2.7376e-06, 4.3202e-06, 3.2837e-06, 4.2007e-06,\n","        3.6579e-06, 2.8820e-06, 3.8836e-06, 3.8082e-06, 3.6063e-06, 3.0047e-06,\n","        3.5454e-06, 4.0272e-06, 3.3097e-06, 4.7001e-06, 4.1901e-06, 3.8059e-06,\n","        3.5720e-06, 2.7784e-06, 3.3791e-06, 3.4935e-06, 3.7773e-06, 3.0843e-06,\n","        2.9892e-06, 3.4121e-06, 3.2927e-06, 4.0600e-06, 4.2092e-06, 3.2147e-06,\n","        3.4320e-06, 3.9439e-06, 3.0362e-06, 2.8972e-06, 3.2261e-06, 2.8795e-06,\n","        3.9268e-06, 3.4671e-06, 3.1633e-06, 3.0530e-06, 3.2056e-06, 2.9723e-06,\n","        3.2206e-06, 2.7608e-06, 2.8831e-06, 3.9470e-06, 3.1275e-06, 3.7836e-06,\n","        3.5576e-06, 2.5605e-06, 3.9732e-06, 3.3790e-06, 3.7884e-06, 3.6959e-06,\n","        3.9634e-06, 4.1439e-06, 3.5668e-06, 2.8595e-06, 2.9854e-06, 3.6511e-06,\n","        3.1166e-06, 2.8932e-06, 3.0731e-06, 3.1001e-06, 3.5898e-06, 2.7810e-06,\n","        4.0286e-06, 3.7737e-06, 2.7998e-06, 3.3732e-06, 4.8843e-06, 2.9459e-06,\n","        3.8628e-06, 2.8062e-06, 3.2376e-06, 3.3460e-06, 3.7254e-06, 3.2094e-06,\n","        3.2103e-06, 3.4281e-06, 3.1815e-06, 2.6341e-06, 3.1273e-06, 3.1072e-06,\n","        3.7130e-06, 3.1912e-06, 2.5562e-06, 3.7426e-06, 2.5143e-06, 3.5638e-06,\n","        4.0744e-06, 3.8129e-06, 3.5549e-06, 2.4478e-06, 2.9129e-06, 3.5248e-06,\n","        2.8423e-06, 3.2883e-06, 3.2964e-06, 3.7055e-06, 3.5606e-06, 3.7185e-06,\n","        3.7010e-06, 3.8800e-06, 3.8743e-06, 3.5569e-06, 2.8312e-06, 3.6435e-06,\n","        3.5279e-06, 2.5042e-06, 2.9706e-06, 4.0584e-06, 4.7154e-06, 3.0725e-06,\n","        4.0585e-06, 4.2784e-06, 3.2445e-06, 3.8040e-06, 3.4227e-06, 3.5416e-06,\n","        3.9879e-06, 3.7906e-06, 3.4846e-06, 3.2291e-06, 2.3418e-06, 3.2221e-06,\n","        3.5253e-06, 4.2171e-06, 4.2531e-06, 3.6599e-06, 3.3163e-06, 4.5837e-06,\n","        3.0191e-06, 3.4707e-06, 3.5156e-06, 3.6137e-06, 3.5928e-06, 4.4261e-06,\n","        3.6464e-06, 4.0039e-06, 3.7821e-06, 2.8654e-06, 3.3492e-06, 3.3694e-06,\n","        2.9422e-06, 3.0833e-06, 2.6204e-06, 3.2160e-06, 3.5257e-06, 3.4174e-06,\n","        3.9552e-06, 3.0422e-06, 2.9097e-06, 2.8426e-06, 3.8297e-06, 4.1898e-06,\n","        3.4371e-06, 2.9013e-06, 3.6215e-06, 2.6134e-06, 3.0280e-06, 4.0913e-06,\n","        4.1387e-06, 2.9269e-06, 3.6951e-06, 3.7556e-06, 3.4491e-06, 2.9259e-06,\n","        3.4926e-06, 3.0437e-06, 2.4309e-06, 3.9658e-06, 2.9546e-06, 2.9295e-06,\n","        3.6483e-06, 3.2714e-06, 3.7057e-06, 3.0474e-06, 3.3712e-06, 3.4423e-06,\n","        4.3340e-06, 3.0396e-06, 4.0254e-06, 3.2629e-06, 4.1385e-06, 3.4532e-06,\n","        3.4873e-06, 3.1361e-06, 2.4949e-06, 3.5222e-06, 3.0918e-06, 4.1808e-06,\n","        3.9109e-06, 3.9713e-06, 2.7363e-06, 3.7999e-06, 3.9773e-06, 3.3033e-06,\n","        3.3785e-06, 3.5867e-06, 3.3114e-06, 3.7970e-06, 3.0541e-06, 4.0319e-06,\n","        4.4321e-06, 3.7566e-06, 3.4352e-06, 3.2626e-06, 2.9697e-06, 3.7853e-06,\n","        3.6291e-06, 3.5598e-06, 3.7253e-06, 3.0914e-06, 3.4763e-06, 3.5591e-06,\n","        3.4552e-06, 3.0872e-06, 3.2357e-06, 2.2510e-06, 3.2364e-06, 3.9133e-06,\n","        4.3136e-06, 3.5791e-06, 3.2667e-06, 3.7940e-06, 4.5541e-06, 3.0240e-06,\n","        3.8363e-06, 3.1751e-06, 3.0650e-06, 3.7552e-06, 2.9450e-06, 3.2804e-06,\n","        2.9778e-06, 3.4686e-06, 2.9281e-06, 3.0828e-06, 2.9266e-06, 2.7001e-06,\n","        3.6554e-06, 3.6490e-06, 3.3763e-06, 2.6730e-06, 3.9095e-06, 3.1979e-06,\n","        2.7070e-06, 3.5943e-06, 3.7404e-06, 2.7763e-06, 3.3181e-06, 3.5812e-06,\n","        3.0268e-06, 3.2236e-06, 3.8938e-06, 3.8582e-06, 2.7810e-06, 3.4735e-06,\n","        3.6473e-06, 2.8811e-06, 3.0341e-06, 4.0511e-06, 3.1570e-06, 3.1812e-06,\n","        3.3824e-06, 4.1424e-06, 3.5307e-06, 3.0381e-06, 3.7151e-06, 3.8561e-06,\n","        3.5662e-06, 3.1081e-06, 3.9430e-06, 2.7726e-06, 2.8552e-06, 3.8173e-06,\n","        3.2599e-06, 3.8184e-06, 3.9250e-06, 4.9812e-06, 4.3726e-06, 3.8076e-06,\n","        3.4521e-06, 4.7052e-06, 3.7173e-06, 2.3566e-06, 3.5966e-06, 3.0181e-06,\n","        3.0299e-06, 3.5167e-06, 4.3647e-06, 2.7246e-06, 3.4304e-06, 3.0998e-06,\n","        3.2561e-06, 2.8351e-06, 3.7865e-06, 3.6895e-06, 3.3357e-06, 3.4307e-06,\n","        4.2474e-06, 2.6177e-06, 2.9974e-06, 3.6425e-06, 3.2868e-06, 3.4651e-06,\n","        3.7242e-06, 3.3770e-06, 3.3211e-06, 3.9115e-06, 3.0708e-06, 2.9978e-06,\n","        3.6335e-06, 4.2891e-06, 2.4903e-06, 3.1452e-06, 3.9572e-06, 3.4782e-06,\n","        2.9131e-06, 3.3413e-06, 3.2399e-06, 3.4667e-06, 3.0955e-06, 3.7132e-06,\n","        3.6978e-06, 3.0773e-06, 3.2766e-06, 3.7202e-06, 3.7883e-06, 2.7131e-06,\n","        3.2639e-06, 3.8775e-06, 3.0836e-06, 3.4331e-06, 3.6607e-06, 3.6053e-06,\n","        4.2589e-06, 3.7643e-06, 3.0530e-06, 4.0696e-06, 3.3291e-06, 3.1676e-06,\n","        2.8736e-06, 3.8927e-06, 3.6436e-06, 4.0191e-06, 4.2023e-06, 3.4597e-06,\n","        3.4202e-06, 3.4762e-06, 3.7112e-06, 3.7527e-06, 3.2278e-06, 3.6658e-06,\n","        2.6233e-06, 3.6471e-06, 3.9274e-06, 2.6799e-06, 3.0841e-06, 3.0535e-06,\n","        3.5484e-06, 3.5417e-06, 3.4461e-06, 3.1870e-06, 2.7746e-06, 3.5826e-06,\n","        3.4027e-06, 4.3029e-06, 3.1700e-06, 4.7033e-06, 3.3174e-06, 3.2492e-06,\n","        4.0524e-06, 2.7269e-06, 3.8852e-06, 3.4369e-06, 2.4207e-06, 3.6519e-06,\n","        3.2736e-06, 3.0410e-06, 4.0857e-06, 3.3603e-06, 2.8370e-06, 3.7870e-06,\n","        4.1910e-06, 3.8454e-06, 2.3356e-06, 3.3284e-06, 3.6577e-06, 2.9866e-06,\n","        3.9979e-06, 3.0734e-06, 2.4290e-06, 3.2677e-06, 3.4650e-06, 3.3273e-06,\n","        2.8793e-06, 3.6631e-06, 4.0471e-06, 3.6379e-06, 4.3486e-06, 2.9731e-06,\n","        3.1857e-06, 3.1845e-06, 4.2362e-06, 3.4869e-06, 3.2202e-06, 3.4629e-06,\n","        3.0343e-06, 3.4186e-06, 4.2757e-06, 3.4790e-06, 2.9842e-06, 3.0346e-06,\n","        3.4356e-06, 3.6381e-06, 3.4559e-06, 2.7121e-06, 3.4752e-06, 3.2668e-06,\n","        4.3141e-06, 4.0307e-06, 3.1212e-06, 3.2636e-06, 2.4575e-06, 3.3505e-06,\n","        3.7012e-06, 3.9449e-06, 3.2345e-06, 3.7245e-06, 2.9436e-06, 4.2416e-06,\n","        3.1297e-06, 3.5032e-06, 3.4341e-06, 3.0093e-06, 2.9583e-06, 3.4758e-06,\n","        2.8267e-06, 3.7297e-06, 4.0600e-06, 3.1525e-06, 3.9847e-06, 4.7734e-06,\n","        3.4056e-06, 2.7571e-06, 2.9794e-06, 4.3323e-06, 2.6165e-06, 3.5603e-06,\n","        2.7290e-06, 3.1363e-06, 3.5435e-06, 3.6903e-06, 3.4453e-06, 3.2796e-06,\n","        3.0735e-06, 3.1046e-06, 3.1403e-06, 3.7511e-06, 3.7738e-06, 4.4271e-06,\n","        2.8356e-06, 3.7479e-06, 2.4245e-06, 4.1133e-06, 3.5979e-06, 3.9161e-06,\n","        3.3026e-06, 3.7681e-06, 4.6810e-06, 3.3863e-06, 4.0259e-06, 3.5317e-06,\n","        2.2759e-06, 3.9236e-06, 3.0395e-06, 3.1035e-06, 3.4747e-06, 3.7639e-06,\n","        3.2772e-06, 3.6144e-06, 3.9809e-06, 3.3319e-06, 3.9061e-06, 3.5473e-06,\n","        4.9118e-06, 3.9436e-06, 3.6731e-06, 3.3776e-06, 3.7663e-06, 3.2141e-06,\n","        2.9616e-06, 3.4257e-06, 3.4670e-06, 3.9819e-06, 4.3617e-06, 3.4337e-06,\n","        3.2629e-06, 3.4480e-06, 3.2391e-06, 3.3131e-06, 3.6889e-06, 5.1628e-06,\n","        3.6832e-06, 2.9469e-06, 3.0793e-06, 3.5883e-06, 3.1325e-06, 2.8149e-06,\n","        2.8348e-06, 3.3993e-06, 3.8614e-06, 4.1078e-06, 3.6081e-06, 3.4591e-06,\n","        3.6931e-06, 3.2396e-06, 3.5936e-06, 4.4563e-06, 3.5176e-06, 3.6496e-06,\n","        4.1486e-06, 3.5809e-06, 3.8821e-06, 3.5019e-06, 3.8211e-06, 2.5272e-06,\n","        3.4011e-06, 3.0663e-06, 3.5794e-06, 3.1775e-06, 3.0860e-06, 2.8729e-06,\n","        3.2103e-06, 4.1998e-06, 4.0193e-06, 3.2450e-06, 3.7957e-06, 3.0986e-06,\n","        3.7923e-06, 3.0257e-06, 4.7247e-06, 3.8204e-06, 4.1671e-06, 4.3611e-06,\n","        2.6525e-06, 3.0798e-06, 2.6772e-06, 3.6263e-06, 4.1240e-06, 4.0678e-06,\n","        3.1507e-06, 3.3923e-06, 3.3423e-06, 3.3834e-06, 4.0201e-06, 3.4740e-06,\n","        3.9926e-06, 3.0865e-06, 3.4183e-06, 2.5805e-06, 4.2242e-06, 3.2656e-06,\n","        3.7381e-06, 4.7393e-06, 3.5132e-06, 2.9118e-06, 3.2319e-06, 3.9029e-06,\n","        2.6141e-06, 3.5531e-06, 3.6830e-06, 2.8836e-06, 3.3334e-06, 3.3403e-06,\n","        4.2207e-06, 3.8197e-06, 2.8405e-06, 3.8032e-06, 3.5929e-06, 2.5003e-06,\n","        2.7994e-06, 3.3380e-06, 3.0038e-06, 2.9183e-06, 2.8607e-06, 3.1073e-06,\n","        3.2434e-06, 3.2268e-06, 3.7712e-06, 3.2705e-06, 3.4862e-06, 3.9797e-06,\n","        3.6382e-06, 3.1539e-06, 3.4690e-06, 2.7816e-06, 4.0556e-06, 3.3388e-06,\n","        3.0954e-06, 3.0459e-06, 3.8578e-06, 2.5370e-06, 3.1959e-06, 3.0926e-06,\n","        3.2490e-06, 4.2769e-06, 3.0853e-06, 2.8465e-06, 3.6932e-06, 2.4953e-06,\n","        3.1531e-06, 3.6165e-06, 2.6308e-06, 2.7427e-06, 3.6300e-06, 3.3732e-06,\n","        3.4093e-06, 3.3347e-06, 2.8952e-06, 3.9268e-06, 3.2831e-06, 3.1986e-06,\n","        3.0416e-06, 3.1895e-06, 3.3644e-06, 3.0686e-06, 3.7013e-06, 4.0231e-06,\n","        4.1533e-06, 2.4814e-06, 4.0271e-06, 3.8591e-06, 4.2927e-06, 2.8482e-06,\n","        3.1292e-06, 3.5592e-06, 3.1643e-06, 2.4748e-06, 3.2370e-06, 3.9019e-06,\n","        3.9160e-06, 3.6440e-06, 3.5450e-06, 3.2997e-06, 3.3443e-06, 3.3877e-06,\n","        3.5583e-06, 3.0427e-06, 3.2133e-06, 3.1281e-06, 3.7815e-06, 4.5032e-06,\n","        3.4522e-06, 3.8715e-06, 3.7111e-06, 4.0258e-06, 3.5426e-06, 3.4234e-06,\n","        3.9453e-06, 3.5818e-06, 3.3793e-06, 3.0571e-06, 3.7196e-06, 2.5264e-06,\n","        4.2230e-06, 5.2402e-06, 2.9107e-06, 3.1522e-06, 4.2125e-06, 3.8465e-06,\n","        3.2203e-06, 2.8024e-06, 3.2769e-06, 4.0822e-06, 2.6785e-06, 3.5943e-06,\n","        3.1769e-06, 3.0323e-06, 3.1505e-06, 3.3604e-06, 2.6856e-06, 2.6099e-06,\n","        3.6354e-06, 2.9039e-06, 3.2915e-06, 3.9039e-06, 3.7986e-06, 3.9896e-06,\n","        3.5563e-06, 3.6424e-06, 3.6492e-06, 3.7722e-06, 4.0737e-06, 4.2996e-06,\n","        3.4050e-06, 3.4039e-06, 3.4994e-06, 2.5366e-06, 2.9664e-06, 3.3244e-06,\n","        3.7679e-06, 3.7230e-06, 3.9992e-06, 3.0332e-06, 3.2463e-06, 3.1703e-06,\n","        3.8758e-06, 3.2723e-06, 2.4055e-06, 2.8930e-06, 3.5653e-06, 2.7827e-06,\n","        3.6225e-06, 3.5991e-06, 2.7071e-06, 3.2843e-06, 4.0672e-06, 3.4398e-06,\n","        3.0583e-06, 3.7854e-06, 4.1163e-06, 4.1671e-06, 3.9490e-06, 3.5257e-06,\n","        4.3669e-06, 3.8251e-06, 3.5466e-06, 3.2557e-06, 2.6757e-06, 3.1999e-06,\n","        4.0886e-06, 3.6410e-06, 3.9961e-06, 3.7807e-06, 3.5408e-06, 4.4198e-06,\n","        3.8675e-06, 4.1456e-06, 2.3539e-06, 2.7117e-06, 3.9470e-06, 3.6845e-06,\n","        3.0310e-06, 3.5168e-06, 4.3907e-06, 2.6711e-06, 4.1277e-06, 3.4724e-06,\n","        3.8705e-06, 2.3887e-06, 3.9755e-06, 3.7400e-06, 3.9664e-06, 3.7097e-06,\n","        3.9358e-06, 3.3192e-06, 4.2311e-06, 4.2727e-06, 3.7065e-06, 4.2151e-06,\n","        3.1205e-06, 3.6341e-06, 3.0463e-06, 4.0245e-06, 3.0137e-06, 3.4534e-06,\n","        3.3277e-06, 2.6536e-06, 3.6118e-06, 3.0971e-06, 3.6502e-06, 4.1095e-06,\n","        3.3439e-06, 3.3713e-06, 4.0812e-06, 2.8322e-06, 3.6084e-06, 3.7777e-06,\n","        3.3994e-06, 3.7044e-06, 3.2359e-06, 2.9566e-06], device='mps:0')}, 6: {'step': tensor(1025.), 'exp_avg': tensor([[ 2.2929e-04, -4.8712e-05, -1.9380e-04,  ...,  3.3300e-05,\n","          3.8167e-06,  1.1348e-03],\n","        [ 3.9421e-04,  3.8582e-05,  7.5873e-04,  ..., -3.0711e-05,\n","         -1.1633e-05,  9.0839e-04],\n","        [ 7.0911e-04,  1.3450e-05,  6.3035e-05,  ..., -8.4667e-06,\n","         -2.3210e-06,  6.3520e-04],\n","        ...,\n","        [ 1.0155e-05, -9.4876e-05,  2.8042e-03,  ...,  7.3400e-05,\n","          1.2170e-05, -2.3579e-03],\n","        [ 1.4761e-05,  9.1032e-05, -2.6120e-03,  ..., -1.8242e-05,\n","         -1.0146e-06, -7.7814e-04],\n","        [ 5.7531e-04,  5.6609e-05,  7.7425e-05,  ..., -3.1358e-05,\n","         -5.7281e-06,  2.3557e-04]], device='mps:0'), 'exp_avg_sq': tensor([[9.7851e-06, 3.5643e-05, 1.2312e-05,  ..., 1.7076e-05, 1.5500e-05,\n","         1.3487e-05],\n","        [4.7391e-05, 1.0025e-05, 1.6728e-05,  ..., 2.9888e-05, 1.9044e-05,\n","         3.3668e-05],\n","        [1.0080e-05, 1.4619e-05, 4.7287e-06,  ..., 9.6396e-06, 1.0469e-05,\n","         5.1170e-05],\n","        ...,\n","        [3.4068e-05, 2.8134e-05, 3.3231e-05,  ..., 3.5567e-05, 5.7019e-05,\n","         4.1407e-05],\n","        [4.4482e-05, 4.3048e-05, 5.9322e-05,  ..., 8.4755e-05, 2.9603e-05,\n","         4.1419e-05],\n","        [3.3720e-05, 4.4000e-05, 1.2939e-05,  ..., 1.8322e-05, 3.0924e-05,\n","         1.7813e-05]], device='mps:0')}, 7: {'step': tensor(1025.), 'exp_avg': tensor([-9.1851e-04,  8.3294e-04,  2.2542e-04,  3.5290e-03,  3.3851e-04,\n","        -2.0144e-03, -1.8806e-03,  1.0821e-03,  3.2936e-03,  4.7734e-04,\n","         1.0417e-03,  9.3609e-04,  6.9147e-04,  7.0716e-04,  1.1445e-03,\n","         2.4297e-04, -1.7696e-03,  1.5645e-03,  1.3209e-03,  1.4441e-04,\n","         5.1022e-04, -5.8374e-04,  2.1032e-04, -3.2605e-03,  2.1524e-04,\n","        -1.3457e-03,  1.1243e-03, -2.5780e-04,  1.4222e-03,  5.3859e-04,\n","        -7.5433e-04, -1.1789e-03, -1.1415e-03, -8.2586e-04, -9.0529e-04,\n","         4.3174e-04, -1.5014e-03,  6.0313e-05, -1.0337e-03,  1.0536e-03,\n","         9.4969e-04,  2.0073e-06, -5.7485e-04,  4.0876e-05,  6.0854e-04,\n","        -6.6863e-04,  4.3152e-04, -7.2623e-04,  1.8113e-03,  3.3022e-03,\n","        -1.3578e-03,  1.9525e-03,  7.1090e-04, -1.1583e-03, -8.3614e-04,\n","         1.4856e-03,  3.2162e-04,  8.0274e-04,  1.5358e-04,  6.0050e-04,\n","         5.8860e-04, -1.5907e-03, -3.6099e-04,  5.6199e-04,  2.2032e-03,\n","        -7.0273e-04,  3.0783e-04, -2.5101e-03, -5.0959e-04, -9.8340e-04,\n","         6.6166e-04, -2.9037e-04, -5.3117e-03,  1.5873e-03, -1.4597e-03,\n","         4.4398e-04,  1.6237e-03,  5.1523e-04,  7.0509e-04,  1.6677e-04,\n","        -7.9819e-04,  2.0531e-04, -7.3534e-04, -1.6333e-03, -3.3403e-03,\n","         7.7224e-04,  9.2231e-04, -1.0953e-03,  8.5649e-04,  1.3147e-03,\n","        -6.7463e-04,  7.8639e-04,  2.2284e-03,  9.5225e-04, -4.8987e-03,\n","        -8.6112e-05, -7.0373e-04, -2.8236e-03,  1.9868e-03, -2.3681e-03,\n","         7.9426e-04,  1.0733e-03], device='mps:0'), 'exp_avg_sq': tensor([2.0296e-05, 1.8178e-05, 1.5100e-05, 3.9611e-05, 1.1607e-05, 5.3793e-05,\n","        3.9423e-05, 1.5649e-05, 2.5415e-05, 2.0555e-05, 3.1492e-05, 2.5810e-05,\n","        3.9907e-05, 2.9364e-05, 1.9304e-05, 4.4548e-05, 2.0118e-05, 2.5917e-05,\n","        2.6837e-05, 1.7527e-05, 1.3656e-05, 2.7374e-05, 1.3927e-05, 3.1853e-05,\n","        2.7709e-05, 5.3319e-05, 4.2525e-05, 4.0204e-05, 3.2018e-05, 1.5278e-05,\n","        1.9437e-05, 1.0337e-05, 3.4650e-05, 1.6937e-05, 2.1888e-05, 3.0477e-05,\n","        6.2723e-05, 4.5243e-05, 5.5636e-05, 3.3833e-05, 9.9638e-05, 1.8253e-05,\n","        2.6510e-05, 4.4661e-05, 1.0327e-05, 3.4009e-05, 1.6599e-05, 3.1788e-05,\n","        4.4028e-05, 1.2983e-04, 3.2065e-05, 3.8378e-05, 1.8729e-05, 3.3634e-05,\n","        2.0192e-05, 3.2628e-05, 1.5739e-05, 1.2214e-05, 1.2671e-05, 1.0737e-05,\n","        8.8252e-06, 3.6448e-05, 9.1249e-06, 8.6534e-06, 3.1871e-05, 1.5077e-05,\n","        2.5085e-05, 3.5223e-05, 2.1602e-05, 2.5404e-05, 2.0719e-05, 1.5762e-05,\n","        7.0038e-05, 5.5927e-05, 8.9284e-05, 4.5027e-05, 4.5965e-05, 4.4607e-05,\n","        4.3346e-05, 1.1595e-05, 2.0820e-05, 3.3350e-05, 3.7451e-05, 4.6255e-05,\n","        8.2174e-05, 6.3846e-05, 2.4385e-05, 2.6248e-05, 2.9153e-05, 9.3973e-05,\n","        6.6014e-05, 2.8141e-05, 6.1108e-05, 3.3259e-05, 3.4116e-05, 2.5665e-05,\n","        3.6737e-05, 6.9788e-05, 1.0188e-04, 5.2776e-05, 5.1913e-05, 2.1803e-05],\n","       device='mps:0')}}, 'param_groups': [{'lr': 0.001, 'betas': (0.9, 0.999), 'eps': 1e-08, 'weight_decay': 0, 'amsgrad': False, 'maximize': False, 'foreach': None, 'capturable': False, 'differentiable': False, 'fused': None, 'params': [0, 1, 2, 3, 4, 5, 6, 7]}]}\n"]}],"source":["print(optimizer.state_dict())\n"]},{"cell_type":"code","execution_count":13,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:34:25.638212Z","iopub.status.busy":"2023-12-30T11:34:25.637136Z","iopub.status.idle":"2023-12-30T11:34:25.956354Z","shell.execute_reply":"2023-12-30T11:34:25.954952Z","shell.execute_reply.started":"2023-12-30T11:34:25.638168Z"},"trusted":true},"outputs":[{"data":{"text/plain":["Text(0.5, 1.0, 'Training and Validation losses')"]},"execution_count":13,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"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","text/plain":["<Figure size 640x480 with 1 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":["plt.plot(train_losses, label='train_losses')\n","plt.plot(valid_losses, label = 'valid_losses')\n","plt.legend(frameon = True)\n","plt.title('Training and Validation losses')"]},{"cell_type":"markdown","metadata":{},"source":["## Testing your network\n","\n","It's good practice to test your trained network on test data, images the network has never seen either in training or validation. This will give you a good estimate for the model's performance on completely new images. Run the test images through the network and measure the accuracy, the same way you did validation. You should be able to reach around 70% accuracy on the test set if the model has been trained well."]},{"cell_type":"code","execution_count":14,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:38:57.538607Z","iopub.status.busy":"2023-12-30T11:38:57.537428Z","iopub.status.idle":"2023-12-30T11:38:57.551180Z","shell.execute_reply":"2023-12-30T11:38:57.549755Z","shell.execute_reply.started":"2023-12-30T11:38:57.538556Z"},"trusted":true},"outputs":[],"source":["# TODO: Do validation on the test set\n","def test_model(model, epochs,testloaders, criterion, optimizer, scheduler, gpu = 'gpu'):\n","    if gpu == 'gpu':\n","        device = 'mps' if torch.backends.mps.is_available() else 'cuda'\n","    else:\n","        device = 'cpu'\n","    print('device is :', device)\n","    model.to(device)\n","    \n","    for e in range(epochs):\n","        running_accuracy_per_epoch = 0\n","        running_test_loss = 0\n","        \n","        model.eval()\n","        for images, labels in tqdm(testloaders):\n","            \n","            images, labels = images.to(device), labels.to(device)\n","            prob_per_batch = torch.exp(model(images))\n","            test_loss = criterion(model(images), labels)\n","            running_test_loss += test_loss.item()\n","        \n","            top_prob_pre_batch ,top_class_per_batch = prob_per_batch.topk(1, dim =1)\n","            truth_val = top_class_per_batch == labels.view(*top_class_per_batch.shape)#this is bytes \n","            running_accuracy_per_epoch += torch.mean(truth_val.type(torch.FloatTensor))\n","    \n","        epoch_test_accuracy = running_accuracy_per_epoch/len(testloaders) # divide by the number of batches to get average accuracy for the epoch\n","        test_loss_per_epoch = running_test_loss/len(testloaders)\n","        test_losses.append(test_loss_per_epoch)\n","        test_acc.append(epoch_test_accuracy)\n","    \n","        print(f'epoch: {e+1}/{epochs}')\n","        print(f'test_loss_per_epoch :{test_losses[-1]}')                     \n","        print(f'epoch_test_accuracy: {epoch_test_accuracy.item()*100}%')# changes single element tensor to a python number\n","        \n","    return test_losses, test_acc"]},{"cell_type":"code","execution_count":15,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:39:00.469581Z","iopub.status.busy":"2023-12-30T11:39:00.468441Z","iopub.status.idle":"2023-12-30T11:39:59.409938Z","shell.execute_reply":"2023-12-30T11:39:59.408845Z","shell.execute_reply.started":"2023-12-30T11:39:00.469533Z"},"trusted":true},"outputs":[{"name":"stdout","output_type":"stream","text":["device is : mps\n"]},{"name":"stderr","output_type":"stream","text":["100%|██████████| 52/52 [00:22<00:00,  2.35it/s]\n"]},{"name":"stdout","output_type":"stream","text":["epoch: 1/5\n","test_loss_per_epoch :0.39198150996190423\n","epoch_test_accuracy: 89.78365659713745%\n"]},{"name":"stderr","output_type":"stream","text":["100%|██████████| 52/52 [00:19<00:00,  2.68it/s]\n"]},{"name":"stdout","output_type":"stream","text":["epoch: 2/5\n","test_loss_per_epoch :0.3620274068089202\n","epoch_test_accuracy: 90.38461446762085%\n"]},{"name":"stderr","output_type":"stream","text":["100%|██████████| 52/52 [00:19<00:00,  2.69it/s]\n"]},{"name":"stdout","output_type":"stream","text":["epoch: 3/5\n","test_loss_per_epoch :0.42866154342818147\n","epoch_test_accuracy: 90.98557829856873%\n"]},{"name":"stderr","output_type":"stream","text":["100%|██████████| 52/52 [00:20<00:00,  2.59it/s]\n"]},{"name":"stdout","output_type":"stream","text":["epoch: 4/5\n","test_loss_per_epoch :0.42576497732303464\n","epoch_test_accuracy: 89.58333134651184%\n"]},{"name":"stderr","output_type":"stream","text":["100%|██████████| 52/52 [00:19<00:00,  2.66it/s]"]},{"name":"stdout","output_type":"stream","text":["epoch: 5/5\n","test_loss_per_epoch :0.4229653222271456\n","epoch_test_accuracy: 91.30609035491943%\n"]},{"name":"stderr","output_type":"stream","text":["\n"]}],"source":["test_losses, test_accuracy = test_model(\n","    new_model, epochs, testloaders, criterion, optimizer, scheduler,'gpu' )"]},{"cell_type":"code","execution_count":16,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:40:01.913577Z","iopub.status.busy":"2023-12-30T11:40:01.912669Z","iopub.status.idle":"2023-12-30T11:40:02.186953Z","shell.execute_reply":"2023-12-30T11:40:02.185561Z","shell.execute_reply.started":"2023-12-30T11:40:01.913532Z"},"trusted":true},"outputs":[{"data":{"image/png":"iVBORw0KGgoAAAANSUhEUgAAAiMAAAGzCAYAAAD9pBdvAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB0n0lEQVR4nO3deVxU5f4H8M+ZgZlh3xlAEVRcUBEVgdRKS9z1umSaenOp7FaWeW3TW7mW/m62WC7XssU2zVzbzFLUNDPABVfcUVBZZd9mYOb8/hgYGVkHgcPA5/16nZfMmbN8DzPjfHme7/McQRRFEUREREQSkUkdABEREbVsTEaIiIhIUkxGiIiISFJMRoiIiEhSTEaIiIhIUkxGiIiISFJMRoiIiEhSTEaIiIhIUkxGiIiISFJMRqhJmj59Ovz9/eu076JFiyAIQv0G1MQIgoBFixYZH2/YsAGCIODatWs17uvv74/p06fXazz38npZsgMHDkAQBBw4cEDqUIgsGpMRMosgCLVa+J+zwezZsyEIAi5fvlzlNq+//joEQcCpU6caMTLz3bp1C4sWLUJsbKzUoViUxvzMFBQUYNGiRbU+VlkytXXr1ns+N9G9sJI6ALIsX3/9tcnjr776Cnv27KmwPjAw8J7Os379euj1+jrt+8Ybb2DevHn3dP76MmXKFKxatQobN27EggULKt1m06ZNCAoKQvfu3et8nscffxyPPfYYlEplnY9Rk1u3bmHx4sXw9/dHjx49TJ67l9eruWuszwxgSEYWL14MABgwYMA9H4+osTAZIbP885//NHn8999/Y8+ePRXW362goAC2tra1Po+1tXWd4gMAKysrWFk1jbd2eHg4AgICsGnTpkqTkSNHjiA+Ph7/93//d0/nkcvlkMvl93SMe3Evr1dzV9fPDFFLwm4aqncDBgxAt27dcOzYMTz44IOwtbXFf/7zHwDADz/8gBEjRsDHxwdKpRLt27fH0qVLodPpTI5xdw3CtWvXIAgC3n33XXzyySdo3749lEolQkNDERMTY7JvZTUjgiDg+eefx86dO9GtWzcolUp07doVu3fvrhD/gQMH0Lt3b6hUKrRv3x4ff/zxPdWhTJkyBefPn8fx48crPLdx40YIgoBJkyZBq9ViwYIFCAkJgZOTE+zs7PDAAw9g//79NZ6jspoRURTx1ltvoXXr1rC1tcVDDz2Es2fPVtg3IyMDL7/8MoKCgmBvbw9HR0cMGzYMJ0+eNPmdhIaGAgBmzJhh7FrYsGEDgMprRvLz8/HSSy/B19cXSqUSnTp1wrvvvou7bxRuzmtTW7V9n5W9V8+dO4eHHnoItra2aNWqFd55550Kx7xx4wbGjBkDOzs7eHp64t///jc0Gk2dYyxPr9dj5cqV6Nq1K1QqFdRqNf71r38hMzPTZLujR49iyJAhcHd3h42NDdq2bYsnnngCgOEz4uHhAQBYvHix8TUqX1tUV1evXsWjjz4KV1dX2Nra4r777sMvv/xSYbtVq1aha9eusLW1hYuLC3r37o2NGzcan8/NzcWcOXPg7+8PpVIJT09PDBo0qMJnIyoqCkOHDoWTkxNsbW3Rv39/HD582GSb2h6LLEPT+PORmp3bt29j2LBheOyxx/DPf/4TarUagOFL097eHnPnzoW9vT327duHBQsWICcnBytWrKjxuBs3bkRubi7+9a9/QRAEvPPOOxg3bhyuXr1a41/nf/75J7Zv347nnnsODg4O+Oijj/DII48gISEBbm5uAIATJ05g6NCh8Pb2xuLFi6HT6bBkyRLjf/J1MWXKFCxevBgbN25Er169jOt1Oh2+//57PPDAA2jTpg3S09Px6aefYtKkSZg5cyZyc3Px2WefYciQIYiOjq7QNVKTBQsW4K233sLw4cMxfPhwHD9+HIMHD4ZWqzXZ7urVq9i5cyceffRRtG3bFikpKfj444/Rv39/nDt3Dj4+PggMDMSSJUuwYMECPP3003jggQcAAH379q303KIo4h//+Af279+PJ598Ej169MBvv/2GV155BTdv3sQHH3xgsn1tXhtzmPM+y8zMxNChQzFu3DhMmDABW7duxWuvvYagoCAMGzYMAFBYWIiBAwciISEBs2fPho+PD77++mvs27fP7Ngq869//QsbNmzAjBkzMHv2bMTHx2P16tU4ceIEDh8+DGtra6SmpmLw4MHw8PDAvHnz4OzsjGvXrmH79u0AAA8PD/zvf//Ds88+i7Fjx2LcuHEAcE/dfwCQkpKCvn37oqCgALNnz4abmxu+/PJL/OMf/8DWrVsxduxYAIauutmzZ2P8+PF48cUXUVRUhFOnTiEqKgqTJ08GADzzzDPYunUrnn/+eXTp0gW3b9/Gn3/+ibi4OONnY9++fRg2bBhCQkKwcOFCyGQyfPHFF3j44Ydx6NAhhIWF1fpYZEFEonswa9Ys8e63Uf/+/UUA4rp16ypsX1BQUGHdv/71L9HW1lYsKioyrps2bZro5+dnfBwfHy8CEN3c3MSMjAzj+h9++EEEIP7000/GdQsXLqwQEwBRoVCIly9fNq47efKkCEBctWqVcd2oUaNEW1tb8ebNm8Z1ly5dEq2srCoc0xyhoaFi69atRZ1OZ1y3e/duEYD48ccfi6IoiiUlJaJGozHZLzMzU1Sr1eITTzxR4XoWLlxofPzFF1+IAMT4+HhRFEUxNTVVVCgU4ogRI0S9Xm/c7j//+Y8IQJw2bZpxXVFRkUlcomj4fSuVSnHJkiXGdTExMSIA8YsvvqhwfXe/Xjt37hQBiG+99ZbJduPHjxcFQTB5HWr72pijtu+zsvfqV199ZVyn0WhELy8v8ZFHHjGuW7lypQhA/P77743r8vPzxYCAABGAuH///lrHdvdn5tChQyIA8dtvvzXZruz9UbZ+x44dIgAxJiamymOnpaVVeG9UZ//+/SIAccuWLVVuM2fOHBGAeOjQIeO63NxcsW3btqK/v7/xvTN69Gixa9eu1Z7PyclJnDVrVpXP6/V6sUOHDuKQIUNM3rcFBQVi27ZtxUGDBtX6WGRZ2E1DDUKpVGLGjBkV1tvY2Bh/zs3NRXp6Oh544AEUFBTg/PnzNR534sSJcHFxMT4u+wv96tWrNe4bERGB9u3bGx93794djo6Oxn11Oh327t2LMWPGwMfHx7hdQECA8S/kuvrnP/+JGzdu4ODBg8Z1GzduhEKhwKOPPgrAUPehUCgAGJrtMzIyUFJSgt69e5vd9Lx3715otVq88MILJt1Lc+bMqbCtUqmETGb4r0Cn0+H27duwt7dHp06d6tzkvWvXLsjlcsyePdtk/UsvvQRRFPHrr7+arK/ptTGXOe8ze3t7k/oNhUKBsLAwk3Pv2rUL3t7eGD9+vHGdra0tnn766TrFV96WLVvg5OSEQYMGIT093biEhITA3t7e2E3n7OwMAPj5559RXFx8z+etrV27diEsLAz333+/cZ29vT2efvppXLt2DefOnTPGd+PGjQrdpuU5OzsjKioKt27dqvT52NhYXLp0CZMnT8bt27eNv4v8/HwMHDgQBw8eNBZK13QssixMRqhBtGrVyvjFWt7Zs2cxduxYODk5wdHRER4eHsYvguzs7BqP26ZNG5PHZYnJ3X3rtdm3bP+yfVNTU1FYWIiAgIAK21W2zhyPPfYY5HK5sf+8qKgIO3bswLBhw0ySqy+//BLdu3eHSqWCm5sbPDw88Msvv9Tqd1Pe9evXAQAdOnQwWe/h4WFyPsCQ+HzwwQfo0KEDlEol3N3d4eHhgVOnTpl93vLn9/HxgYODg8n6shEjZfGVqem1MZc577PWrVtXqAe6+9zXr19HQEBAhe06depUp/jKu3TpErKzs+Hp6QkPDw+TJS8vD6mpqQCA/v3745FHHsHixYvh7u6O0aNH44svvqi3upWqXL9+vdLrvPu1fO2112Bvb4+wsDB06NABs2bNqlDn8c477+DMmTPw9fVFWFgYFi1aZJL0Xbp0CQAwbdq0Cr+LTz/9FBqNxvj61XQssiysGaEGUf4v0zJZWVno378/HB0dsWTJErRv3x4qlQrHjx/Ha6+9VquhoVWNGBHvKoqs733vVVlx3bZt27BmzRr89NNPyM3NxZQpU4zbfPPNN5g+fTrGjBmDV155BZ6enpDL5Vi+fDmuXLnSYLEtW7YMb775Jp544gksXboUrq6ukMlkmDNnTqMN163P18bc95mU7wvAkAx6enri22+/rfT5snqlsvlA/v77b/z000/47bff8MQTT+C9997D33//DXt7+0aJtyqBgYG4cOECfv75Z+zevRvbtm3D2rVrsWDBAuNw4wkTJuCBBx7Ajh078Pvvv2PFihX473//i+3bt2PYsGHG12bFihVV1kiVXWdNxyLLwmSEGs2BAwdw+/ZtbN++HQ8++KBxfXx8vIRR3eHp6QmVSlXpBGXVTVpWW1OmTMHu3bvx66+/YuPGjXB0dMSoUaOMz2/duhXt2rXD9u3bTf4CX7hwodnn8vPzA2D4S7Ndu3bG9WlpaRVaG7Zu3YqHHnoIn332mcn6rKwsuLu7Gx+bM5rIz88Pe/fuRW5urknrSFkXSVl8DaEh3md+fn44c+YMRFE0+T1cuHDhnmIFgPbt22Pv3r3o169fpUn83e677z7cd999ePvtt7Fx40ZMmTIF3333HZ566qkGmXnYz8+v0uus7LW0s7PDxIkTMXHiRGi1WowbNw5vv/025s+fD5VKBQDw9vbGc889h+eeew6pqano1asX3n77bQwbNszYVefo6IiIiIgaY6vuWGRZ2E1DjabsL9Dyf3FqtVqsXbtWqpBMyOVyREREYOfOnSb90JcvX65Q41AXY8aMga2tLdauXYtff/0V48aNM/4HXXZ+wPT3ExUVhSNHjph9roiICFhbW2PVqlUmx1u5cmWFbeVyeYVWgC1btuDmzZsm6+zs7AAYkpSaDB8+HDqdDqtXrzZZ/8EHH0AQhAb9smiI99nw4cNx69Ytk5lKCwoK8Mknn9Q90FITJkyATqfD0qVLKzxXUlJi/H1nZmZWeJ3KWg/KumrK5vKpzWtUW8OHD0d0dLTJ+zA/Px+ffPIJ/P390aVLFwCGEXTlKRQKdOnSBaIoori4GDqdrkIXmaenJ3x8fIzxh4SEoH379nj33XeRl5dXIZa0tDQAqNWxyLKwZYQaTd++feHi4oJp06YZp0n/+uuvG605vDYWLVqE33//Hf369cOzzz5r/ELt1q1bhWnQFy1ahMWLF2P//v21mu3S3t4eY8aMMdaNlO+iAYCRI0di+/btGDt2LEaMGIH4+HisW7cOXbp0qfQ/5up4eHjg5ZdfxvLlyzFy5EgMHz4cJ06cwK+//mrS2lF23iVLlmDGjBno27cvTp8+jW+//dakRQUw/AXv7OyMdevWwcHBAXZ2dggPD0fbtm0rnH/UqFF46KGH8Prrr+PatWsIDg7G77//jh9++AFz5swxKVY1hyAI6N+/f7XTnTfE+2zmzJlYvXo1pk6dimPHjsHb2xtff/21WRP5VaV///7417/+heXLlyM2NhaDBw+GtbU1Ll26hC1btuDDDz/E+PHj8eWXX2Lt2rUYO3Ys2rdvj9zcXKxfvx6Ojo4YPnw4AEP3aJcuXbB582Z07NgRrq6u6NatG7p161ZtDNu2bau0gHzatGmYN28eNm3ahGHDhmH27NlwdXXFl19+ifj4eGzbts1Y/Dx48GB4eXmhX79+UKvViIuLw+rVqzFixAg4ODggKysLrVu3xvjx4xEcHAx7e3vs3bsXMTExeO+99wAAMpkMn376KYYNG4auXbtixowZaNWqFW7evIn9+/fD0dHR2MVZ07HIwkgwgoeakaqG9lY1xO/w4cPifffdJ9rY2Ig+Pj7iq6++Kv72228VhkdWNbR3xYoVFY6Ju4YyVjW0t7JhgH5+fibDXEVRFCMjI8WePXuKCoVCbN++vfjpp5+KL730kqhSqUy2e+mll0RBEMS4uLhKr7Uyv/zyiwhA9Pb2rjCcVq/Xi8uWLRP9/PxEpVIp9uzZU/z5558r/C4qu+a7h/aKoijqdDpx8eLFore3t2hjYyMOGDBAPHPmTIVrLioqEl966SXjdv369ROPHDki9u/fX+zfv7/JeX/44QexS5cuxqHOZcN8K4sxNzdX/Pe//y36+PiI1tbWYocOHcQVK1aYDNksu5bavDa5ubkiAPGxxx6r9HdbXm3fZ1W9Vyu7nuvXr4v/+Mc/RFtbW9Hd3V188cUXjcNv72Vob5lPPvlEDAkJEW1sbEQHBwcxKChIfPXVV8Vbt26JoiiKx48fFydNmiS2adNGVCqVoqenpzhy5Ejx6NGjJsf566+/xJCQEFGhUNQ4zLdsaG9VS9lw3itXrojjx48XnZ2dRZVKJYaFhYk///yzybE+/vhj8cEHHxTd3NxEpVIptm/fXnzllVfE7OxsURQNQ6ZfeeUVMTg4WHRwcBDt7OzE4OBgce3atRXiOnHihDhu3Djjsfz8/MQJEyaIkZGRZh+LLIMgik3oz1KiJmrMmDE4e/assdofAMLCwuDn54ctW7ZIGFnLsWvXLowcORInT55EUFCQ1OEQUT1izQjRXQoLC00eX7p0Cbt27TLpisnJycHJkyexZMmSRo6u5dq/fz8ee+wxJiJEzRBbRoju4u3tjenTp6Ndu3a4fv06/ve//0Gj0eDEiRMV5u0gIqJ7xwJWorsMHToUmzZtQnJyMpRKJfr06YNly5YxESEiaiBsGSEiIiJJsWaEiIiIJMVkhIiIiCRlETUjer0et27dgoODQ4NMd0xERET1TxRF5ObmwsfHxzhBXmUsIhm5desWfH19pQ6DiIiI6iAxMRGtW7eu8nmLSEbKbrSVmJgIR0dHiaMhIiKi2sjJyYGvr6/JDTMrYxHJSFnXjKOjI5MRIiIiC1NTiQULWImIiEhSTEaIiIhIUkxGiIiISFIWUTNCRERNiyiKKCkpgU6nkzoUkpBcLoeVldU9T7vBZISIiMyi1WqRlJSEgoICqUOhJsDW1hbe3t5QKBR1PgaTESIiqjW9Xo/4+HjI5XL4+PhAoVBwMsoWShRFaLVapKWlIT4+Hh06dKh2YrPqMBkhIqJa02q10Ov18PX1ha2trdThkMRsbGxgbW2N69evQ6vVQqVS1ek4LGAlIiKz1fUvYGp+6uO9wHcTERERSYrJCBEREUnK7GTk4MGDGDVqFHx8fCAIAnbu3Fnt9klJSZg8eTI6duwImUyGOXPm1DFUIiKipsHf3x8rV66sl2MdOHAAgiAgKyurXo5nicxORvLz8xEcHIw1a9bUanuNRgMPDw+88cYbCA4ONjtAIiKi+jBgwIB6+4M4JiYGTz/9dL0ci+owmmbYsGEYNmxYrbf39/fHhx9+CAD4/PPPzT1dgzpwIRXf/J2AhaO6wNeVVeFERC2ZKIrQ6XSwsqr5q9HDw6MRImo5mmTNiEajQU5OjsnSED77Mx5741KwKTqhQY5PRNQSiKKIAm1Joy+iKNY6xunTp+OPP/7Ahx9+CEEQIAgCNmzYAEEQ8OuvvyIkJARKpRJ//vknrly5gtGjR0OtVsPe3h6hoaHYu3evyfHu7qYRBAGffvopxo4dC1tbW3To0AE//vhjnX+n27ZtQ9euXaFUKuHv74/33nvP5Pm1a9eiQ4cOUKlUUKvVGD9+vPG5rVu3IigoCDY2NnBzc0NERATy8/ONz3/66acIDAyESqVC586dsXbtWuNzWq0Wzz//PLy9vaFSqeDn54fly5fX+Tpqq0nOM7J8+XIsXry4wc8zOawNDl1Kx/dHb2BOREcorJpkbkZE1KQVFuvQZcFvjX7ec0uGwFZRu6+xDz/8EBcvXkS3bt2wZMkSAMDZs2cBAPPmzcO7776Ldu3awcXFBYmJiRg+fDjefvttKJVKfPXVVxg1ahQuXLiANm3aVHmOxYsX45133sGKFSuwatUqTJkyBdevX4erq6tZ13Xs2DFMmDABixYtwsSJE/HXX3/hueeeg5ubG6ZPn46jR49i9uzZ+Prrr9G3b19kZGTg0KFDAAx1mpMmTcI777yDsWPHIjc3F4cOHTImbt9++y0WLFiA1atXo2fPnjhx4gRmzpwJOzs7TJs2DR999BF+/PFHfP/992jTpg0SExORmJhoVvx10SSTkfnz52Pu3LnGxzk5OfD19a3380R0UcPDQYm0XA32nEvBiO7e9X4OIiKSnpOTExQKBWxtbeHl5QUAOH/+PABgyZIlGDRokHFbV1dXkxrHpUuXYseOHfjxxx/x/PPPV3mO6dOnY9KkSQCAZcuW4aOPPkJ0dDSGDh1qVqzvv/8+Bg4ciDfffBMA0LFjR5w7dw4rVqzA9OnTkZCQADs7O4wcORIODg7w8/NDz549ARiSkZKSEowbNw5+fn4AgKCgIOOxFy5ciPfeew/jxo0DALRt2xbnzp3Dxx9/jGnTpiEhIQEdOnTA/fffD0EQjMdoaE0yGVEqlVAqlQ1+Hmu5DBN6t8aa/VewMfo6kxEiojqwsZbj3JIhkpy3PvTu3dvkcV5eHhYtWoRffvnF+OVeWFiIhITqu/S7d+9u/NnOzg6Ojo5ITU01O564uDiMHj3aZF2/fv2wcuVK6HQ6DBo0CH5+fmjXrh2GDh2KoUOHGruHgoODMXDgQAQFBWHIkCEYPHgwxo8fDxcXF+Tn5+PKlSt48sknMXPmTOOxS0pK4OTkBMCQUA0aNAidOnXC0KFDMXLkSAwePNjsazBXi++XeCy0DQQBOHz5Nq6l59e8AxERmRAEAbYKq0Zf6uueOHZ2diaPX375ZezYsQPLli3DoUOHEBsbi6CgIGi12mqPY21tXeH3otfr6yXG8hwcHHD8+HFs2rQJ3t7eWLBgAYKDg5GVlQW5XI49e/bg119/RZcuXbBq1Sp06tQJ8fHxyMvLAwCsX78esbGxxuXMmTP4+++/AQC9evVCfHw8li5disLCQkyYMMGkHqWhmJ2M5OXlGS8AAOLj4xEbG2vMGOfPn4+pU6ea7FO2fV5eHtLS0hAbG4tz587de/T1wNfVFv07GqqiWchKRNR8KRQK6HS6Grc7fPgwpk+fjrFjxyIoKAheXl64du1awwdYKjAwEIcPH64QU8eOHSGXG1qDrKysEBERgXfeeQenTp3CtWvXsG/fPgCGJKhfv35YvHgxTpw4AYVCgR07dkCtVsPHxwdXr15FQECAydK2bVvjuRwdHTFx4kSsX78emzdvxrZt25CRkdGg12x2N83Ro0fx0EMPGR+X1XZMmzYNGzZsQFJSUoWmrLK+LMBQmLNx40b4+fk16otbnclhbXDgQhq2HLuBuYM7QmlVP01/RETUdPj7+yMqKgrXrl2Dvb19la0WHTp0wPbt2zFq1CgIgoA333yzQVo4qvLSSy8hNDQUS5cuxcSJE3HkyBGsXr3aOOrl559/xtWrV/Hggw/CxcUFu3btgl6vR6dOnRAVFYXIyEgMHjwYnp6eiIqKQlpaGgIDAwEYimxnz54NJycnDB06FBqNBkePHkVmZibmzp2L999/H97e3ujZsydkMhm2bNkCLy8vODs7N+g1m52MDBgwoNrhVBs2bKiwzpzhV1J4uLMnvBxVSM4pwu4zyRjdo5XUIRERUT17+eWXMW3aNHTp0gWFhYX44osvKt3u/fffxxNPPIG+ffvC3d0dr732WoNNMVGZXr164fvvv8eCBQuwdOlSeHt7Y8mSJZg+fToAwNnZGdu3b8eiRYtQVFSEDh06YNOmTejatSvi4uJw8OBBrFy5Ejk5OfDz88N7771nnB/sqaeegq2tLVasWIFXXnkFdnZ2CAoKMk4G5+DggHfeeQeXLl2CXC5HaGgodu3a1eA3RhTEpp4pwDCaxsnJCdnZ2XB0dGyQc7y/5yI+iryE8Lau2PyvPg1yDiIiS1dUVIT4+Hi0bdu2zreLp+aluvdEbb+/W3wBa5nHQn0hE4Co+AxcTs2TOhwiIqIWg8lIKR9nGzzc2RMAC1mJiKj+PPPMM7C3t690eeaZZ6QOr0lokvOMSGVyeBvsjUvFtuM38MqQTlDV0xh2IiJquZYsWYKXX3650ucaqvTA0jAZKad/R0+0crbBzaxC7DqdhHG9WksdEhERWThPT094enpKHUaTxm6acuQyARNDDdPOb4xiVw0REVFjYDJyl4mhvpDLBBy9nomLKblSh0NERNTsMRm5i9pRhYhAQ3MaW0eIiIgaHpORSkwON9ylcNvxGyjU1jx1MBEREdUdk5FKPBDgDl9XG+QWleCnU7ekDoeIiKhZYzJSCZlMwGOhbQCwq4aIiAz8/f2xcuVK42NBELBz584qt7927RoEQTDeWLY6Bw4cgCAIyMrKuuc4LRGTkSpM6O0LK5mA2MQsnLvVePckICIiy5CUlGS85wvdGyYjVfBwUGJIVy8AwMbo6xJHQ0RETY2XlxeUSqXUYTQLTEaqMTnc0FWz88Qt5GtKJI6GiKiJEkVAm9/4ixn3ef3kk0/g4+MDvV5vsn706NF44okncOXKFYwePRpqtRr29vYIDQ3F3r17qz3m3d000dHR6NmzJ1QqFXr37o0TJ06Y9Wu827Zt29C1a1colUr4+/vjvffeM3l+7dq16NChA1QqFdRqNcaPH298buvWrQgKCoKNjQ3c3NwQERGB/Px84/OffvopAgMDoVKp0LlzZ6xdu9b4nFarxfPPPw9vb2+oVCr4+flh+fLl93QtNeEMrNXo084N/m62uHa7AD+evIVJYW2kDomIqOkpLgCW+TT+ef9zC1DY1WrTRx99FC+88AL279+PgQMHAgAyMjKwe/du7Nq1C3l5eRg+fDjefvttKJVKfPXVVxg1ahQuXLiANm1q/r8/Ly8PI0eOxKBBg/DNN98gPj4eL774Yp0v7dixY5gwYQIWLVqEiRMn4q+//sJzzz0HNzc3TJ8+HUePHsXs2bPx9ddfo2/fvsjIyMChQ4cAGLqPJk2ahHfeeQdjx45Fbm4uDh06BLE0efv222+xYMECrF69Gj179sSJEycwc+ZM2NnZYdq0afjoo4/w448/4vvvv0ebNm2QmJiIxMTEOl9LbTAZqYZMJmBSWBss//U8NkYlMBkhIrJQLi4uGDZsGDZu3GhMRrZu3Qp3d3c89NBDkMlkCA4ONm6/dOlS7NixAz/++COef/75Go+/ceNG6PV6fPbZZ1CpVOjatStu3LiBZ599tk7xvv/++xg4cCDefPNNAEDHjh1x7tw5rFixAtOnT0dCQgLs7OwwcuRIODg4wM/PDz179gRgSEZKSkowbtw4+PkZpqoICgoyHnvhwoV47733MG7cOABA27Ztce7cOXz88ceYNm0aEhIS0KFDB9x///0QBMF4jIbEZKQG40Na473fL+L0zWycvpGNoNZOUodERNS0WNsaWimkOK8ZpkyZgpkzZ2Lt2rVQKpX49ttv8dhjj0EmkyEvLw+LFi3CL7/8YvwyLywsREJC7UZUxsXFoXv37lCpVMZ1ffr0MSu+u483evRok3X9+vXDypUrodPpMGjQIPj5+aFdu3YYOnQohg4dirFjx8LW1hbBwcEYOHAggoKCMGTIEAwePBjjx4+Hi4sL8vPzceXKFTz55JOYOXOm8dglJSVwcjJ8v02fPh2DBg1Cp06dMHToUIwcORKDBw+u87XUBmtGauBmr8TQbixkJSKqkiAYuksaexEEs8IcNWoURFHEL7/8gsTERBw6dAhTpkwBALz88svYsWMHli1bhkOHDiE2NhZBQUHQarUN8Ru7Zw4ODjh+/Dg2bdoEb29vLFiwAMHBwcjKyoJcLseePXvw66+/okuXLli1ahU6deqE+Ph45OXlAQDWr1+P2NhY43LmzBn8/fffAIBevXohPj4eS5cuRWFhISZMmGBSj9IQmIzUQlkh6w+xt5BbVCxxNEREVBcqlQrjxo3Dt99+i02bNqFTp07o1asXAODw4cOYPn06xo4di6CgIHh5eeHatWu1PnZgYCBOnTqFoqIi47qyL/e6CAwMxOHDh03WHT58GB07doRcLgcAWFlZISIiAu+88w5OnTqFa9euYd++fQAMxbX9+vXD4sWLceLECSgUCuzYsQNqtRo+Pj64evUqAgICTJa2bdsaz+Xo6IiJEydi/fr12Lx5M7Zt24aMjIw6X09N2E1TC+FtXdHeww5X0vKxM/YWHr+v4fvPiIio/k2ZMgUjR47E2bNn8c9//tO4vkOHDti+fTtGjRoFQRDw5ptvVhh5U53Jkyfj9ddfx8yZMzF//nxcu3YN7777bp3jfOmllxAaGoqlS5di4sSJOHLkCFavXm0c9fLzzz/j6tWrePDBB+Hi4oJdu3ZBr9ejU6dOiIqKQmRkJAYPHgxPT09ERUUhLS0NgYGBAIDFixdj9uzZcHJywtChQ6HRaHD06FFkZmZi7ty5eP/99+Ht7Y2ePXtCJpNhy5Yt8PLygrOzc52vpyZsGakFQRCMxasboxKMFclERGRZHn74Ybi6uuLChQuYPHmycf37778PFxcX9O3bF6NGjcKQIUOMrSa1YW9vj59++gmnT59Gz5498frrr+O///1vnePs1asXvv/+e3z33Xfo1q0bFixYgCVLlmD69OkAAGdnZ2zfvh0PP/wwAgMDsW7dOmzatAldu3aFo6MjDh48iOHDh6Njx45444038N577xknaHvqqafw6aef4osvvkBQUBD69++PDRs2GFtGHBwc8M4776B3794IDQ3FtWvXsGvXLshkDZcyCKIFfLPm5OTAyckJ2dnZcHR0lCSGrAItwpZFQluix47n+qJnGxdJ4iAiklJRURHi4+PRtm1bk2JNarmqe0/U9vubLSO15GyrwMggbwC8Xw0REVF9YjJihrJC1p9O3UJ2IQtZiYiodp555hnY29tXujzzzDNShyc5FrCaIcTPBR3V9riYkocdx29ger+2Ne9EREQt3pIlS/Dyyy9X+pxU5QdNCZMRMwiCgCnhflj441lsjE7AtL7+EMwc505ERC2Pp6cnPD09pQ6jyWI3jZnG9GwFlbUMF1PycOx6ptThEBERWTwmI2ZysrHGqO6GG0KxkJWIiOjeMRmpg7JC1p9PJyGroGlOFUxERGQpmIzUQQ9fZwR6O0JbosfWYzekDoeIiMiiMRmpA0Mha+mMrNGckZWIiOheMBmpo9E9fGCrkONqWj6i4hvu5kFERNQ8XLt2DYIgIDY2VupQmhyzk5GDBw9i1KhR8PHxgSAI2LlzZ437HDhwAL169YJSqURAQAA2bNhQh1CbFgeVNUb3YCErEZGlGDBgAObMmVNvx5s+fTrGjBlTb8drycxORvLz8xEcHIw1a9bUavv4+HiMGDECDz30EGJjYzFnzhw89dRT+O2338wOtqmZHGa4e++vZ5JwO08jcTRERESWyexkZNiwYXjrrbcwduzYWm2/bt06tG3bFu+99x4CAwPx/PPPY/z48fjggw/MDrapCWrthKBWTijWiSxkJaIWSxRFFBQXNPpiTr3e9OnT8ccff+DDDz+EIAgQBAHXrl3DmTNnMGzYMNjb20OtVuPxxx9Henq6cb+tW7ciKCgINjY2cHNzQ0REBPLz87Fo0SJ8+eWX+OGHH4zHO3DggNm/uz/++ANhYWFQKpXw9vbGvHnzUFJSUuP5AUOvQ1hYGOzs7ODs7Ix+/frh+vXrxn1/+OEH9OrVCyqVCu3atcPixYuNxxZFEYsWLUKbNm2gVCrh4+OD2bNnmx1/fWnwGViPHDmCiIgIk3VDhgyptqlMo9FAo7nT0pCTk9NQ4d2zKeFtMG/7aWyKTsDMB9pBJuOMrETUshSWFCJ8Y3ijnzdqchRsrW1rte2HH36Iixcvolu3bliyZAkAwNraGmFhYXjqqafwwQcfoLCwEK+99homTJiAffv2ISkpCZMmTcI777yDsWPHIjc3F4cOHYIoinj55ZcRFxeHnJwcfPHFFwAAV1dXs+K/efMmhg8fjunTp+Orr77C+fPnMXPmTKhUKixatKja85eUlGDMmDGYOXMmNm3aBK1Wi+joaOOs4IcOHcLUqVPx0Ucf4YEHHsCVK1fw9NNPAwAWLlyIbdu24YMPPsB3332Hrl27Ijk5GSdPnjQr/vrU4MlIcnIy1Gq1yTq1Wo2cnBwUFhbCxsamwj7Lly/H4sWLGzq0ejEq2Adv/RKHa7cLcOTqbfQLcJc6JCIiuouTkxMUCgVsbW3h5eUFAHjrrbfQs2dPLFu2zLjd559/Dl9fX1y8eBF5eXkoKSnBuHHj4Odn6JYPCgoybmtjYwONRmM8nrnWrl0LX19frF69GoIgoHPnzrh16xZee+01LFiwAElJSVWePyMjA9nZ2Rg5ciTat28PAAgMDDQee/HixZg3bx6mTZsGAGjXrh2WLl2KV199FQsXLkRCQgK8vLwQEREBa2trtGnTBmFhYXW6jvrQJO9NM3/+fMydO9f4OCcnB76+vhJGVDU7pRXG9PTBN38nYGNUApMRImpxbKxsEDU5SpLz3ouTJ09i//79sLe3r/DclStXMHjwYAwcOBBBQUEYMmQIBg8ejPHjx8PFxeWezlsmLi4Offr0MbnHWb9+/ZCXl4cbN24gODi4yvO7urpi+vTpGDJkCAYNGoSIiAhMmDAB3t7exms7fPgw3n77beOxdTodioqKUFBQgEcffRQrV65Eu3btMHToUAwfPhyjRo2ClZU0aUGDD+318vJCSkqKybqUlBQ4OjpW2ioCAEqlEo6OjiZLU1ZWyPrb2WSk5hZJHA0RUeMSBAG21raNvtzrjUrz8vIwatQoxMbGmiyXLl3Cgw8+CLlcjj179uDXX39Fly5dsGrVKnTq1Anx8fH19JurXk3n/+KLL3DkyBH07dsXmzdvRseOHfH3338br23x4sUm13X69GlcunQJKpUKvr6+uHDhAtauXQsbGxs899xzePDBB1FcXNwo13a3Bk9G+vTpg8jISJN1e/bsQZ8+fRr61I2mi48jevg6o0QvYstRFrISETVFCoUCOp3O+LhXr144e/Ys/P39ERAQYLLY2dkBMCRa/fr1w+LFi3HixAkoFArs2LGj0uOZKzAwEEeOHDEpxD18+DAcHBzQunXrGs8PAD179sT8+fPx119/oVu3bti4caPx2i5cuFDhugICAiCTGb76bWxsMGrUKHz00Uc4cOAAjhw5gtOnT9f5eu6F2clIXl6eMcsCDEN3Y2NjkZBgmGtj/vz5mDp1qnH7Z555BlevXsWrr76K8+fPY+3atfj+++/x73//u36uoIkom5H1u5gE6PWckZWIqKnx9/dHVFQUrl27hvT0dMyaNQsZGRmYNGkSYmJicOXKFfz222+YMWMGdDodoqKisGzZMhw9ehQJCQnYvn070tLSjLUZ/v7+OHXqFC5cuID09HSzWxWee+45JCYm4oUXXsD58+fxww8/YOHChZg7dy5kMlm154+Pj8f8+fNx5MgRXL9+Hb///jsuXbpkjG3BggX46quvsHjxYpw9exZxcXH47rvv8MYbbwAANmzYgM8++wxnzpzB1atX8c0338DGxsZYm9LoRDPt379fBFBhmTZtmiiKojht2jSxf//+Ffbp0aOHqFAoxHbt2olffPGFWefMzs4WAYjZ2dnmhttoCjQlYreFu0W/134WD1xIlTocIqIGUVhYKJ47d04sLCyUOhSzXbhwQbzvvvtEGxsbEYAYHx8vXrx4URw7dqzo7Ows2tjYiJ07dxbnzJkj6vV68dy5c+KQIUNEDw8PUalUih07dhRXrVplPF5qaqo4aNAg0d7eXgQg7t+/v9rzx8fHiwDEEydOGNcdOHBADA0NFRUKhejl5SW+9tprYnFxsSiKYrXnT05OFseMGSN6e3uLCoVC9PPzExcsWCDqdDrjsXfv3i327dtXtLGxER0dHcWwsDDxk08+EUVRFHfs2CGGh4eLjo6Oop2dnXjfffeJe/furdPvtbr3RG2/vwVRbPo3VsnJyYGTkxOys7ObdP3Ioh/PYsNf1zCkqxofP95b6nCIiOpdUVER4uPj0bZtW6hUKqnDoSaguvdEbb+/eW+aejS5tKtmb1wqUnJYyEpERFQbTEbqUUe1A3r7uUCnF7E5JlHqcIiIqBEtW7YM9vb2lS7Dhg2TOrwmrUnOM2LJptzXBkevZ+K76ATMeigAcs7ISkTUIjzzzDOYMGFCpc9VNZUFGTAZqWfDunlj8U/ncCu7CH9cTMXDndU170RERBbP1dXV7CnhyYDdNPVMZS3HI70M48M3RiVIHA0RUcOwgLEP1Ejq473AZKQBTAozFLLuO5+KW1mFEkdDRFR/rK2tAQAFBQUSR0JNRdl7oey9URfspmkAAZ72CG/riqj4DHwXk4i5gzpKHRIRUb2Qy+VwdnZGamoqAMDW9t6nZSfLJIoiCgoKkJqaCmdnZ8jl8jofi8lIA5lynx+i4jOwOSYBsx8OgJWcjVBE1DyU3aW2LCGhls3Z2bnOdy4uw2SkgQzpqoarnQIpORrsO5+KwV3v7YUiImoqBEGAt7c3PD09JbuxGjUN1tbW99QiUobJSANRWsnxaEhrfHzwKjZGJzAZIaJmRy6X18sXERH7DhpQWSHrHxfTkJjBYi8iIqLKMBlpQP7udugX4AZRNNzNl4iIiCpiMtLApoQbbsf8/dEbKNbpJY6GiIio6WEy0sAGdVHD3V6JtFwN9p5LkTocIiKiJofJSAOzlsswoXfpjKzR7KohIiK6G5ORRjAprA0EATh0KR3X0vOlDoeIiKhJYTLSCHxdbfFABw8AwCYWshIREZlgMtJIpoQbhvluPXoD2hIWshIREZVhMtJIBnb2hNpRidv5Wvx2NlnqcIiIiJoMJiONxEouw8TevgCAjVHsqiEiIirDZKQRTQxrA5kAHLl6G1fS8qQOh4iIqElgMtKIWjnbYEAnTwDAJraOEBERAWAy0uiMhazHb6CoWCdxNERERNJjMtLIBnTyhI+TClkFxdh9hoWsRERETEYamVwmYGKooXWEhaxERERMRiQxMdQXcpmA6GsZuJiSK3U4REREkmIyIgEvJxUe7mwoZGXrCBERtXRMRiRSVsi6nYWsRETUwjEZkciDHTzQ2sUGOUUl+PlUktThEBERSYbJiERkMgGTwsoKWa9LHA0REZF0mIxI6NHerWElE3A8IQtxSTlSh0NERCQJJiMS8nRQYXBXNQAWshIRUctVp2RkzZo18Pf3h0qlQnh4OKKjo6vctri4GEuWLEH79u2hUqkQHByM3bt31zng5mZymB8AYOeJmyjQlkgcDRERUeMzOxnZvHkz5s6di4ULF+L48eMIDg7GkCFDkJqaWun2b7zxBj7++GOsWrUK586dwzPPPIOxY8fixIkT9xx8c9C3vRv83GyRqynBTydvSR0OERFRoxNEURTN2SE8PByhoaFYvXo1AECv18PX1xcvvPAC5s2bV2F7Hx8fvP7665g1a5Zx3SOPPAIbGxt88803lZ5Do9FAo9EYH+fk5MDX1xfZ2dlwdHQ0J1yLsO6PK/i/X88juLUTfnj+fqnDISIiqhc5OTlwcnKq8fvbrJYRrVaLY8eOISIi4s4BZDJERETgyJEjle6j0WigUqlM1tnY2ODPP/+s8jzLly+Hk5OTcfH19TUnTIszPqQ1rOUCTt7Ixpmb2VKHQ0RE1KjMSkbS09Oh0+mgVqtN1qvVaiQnV37TtyFDhuD999/HpUuXoNfrsWfPHmzfvh1JSVXPrTF//nxkZ2cbl8TERHPCtDju9koM7eYNAPiWhaxERNTCNPhomg8//BAdOnRA586doVAo8Pzzz2PGjBmQyao+tVKphKOjo8nS3E0unXPkx9ibyNOwkJWIiFoOs5IRd3d3yOVypKSkmKxPSUmBl5dXpft4eHhg586dyM/Px/Xr13H+/HnY29ujXbt2dY+6GbqvnSvaedghX6vDD7E3pQ6HiIio0ZiVjCgUCoSEhCAyMtK4Tq/XIzIyEn369Kl2X5VKhVatWqGkpATbtm3D6NGj6xZxMyUIgrF1ZGNUAsysKyYiIrJYZnfTzJ07F+vXr8eXX36JuLg4PPvss8jPz8eMGTMAAFOnTsX8+fON20dFRWH79u24evUqDh06hKFDh0Kv1+PVV1+tv6toJh7p1RoKKxnO3srByRssZCUiopbBytwdJk6ciLS0NCxYsADJycno0aMHdu/ebSxqTUhIMKkHKSoqwhtvvIGrV6/C3t4ew4cPx9dffw1nZ+d6u4jmwsVOgRFB3thx4iY2Rl1HD19nqUMiIiJqcGbPMyKF2o5Tbg5irmXg0XVHYGMtR9TrA+GospY6JCIiojppkHlGqOH19nNBB097FBbrsPMEC1mJiKj5YzLSxAiCgMnhLGQlIqKWg8lIEzSuZ2sorWQ4n5yL4wmZUodDRETUoJiMNEFOttYYFewDgDOyEhFR88dkpIkq66r55VQSsguKJY6GiIio4TAZaaJ6+jqjs5cDNCV6bDt+Q+pwiIiIGgyTkSZKEARMKStkjWYhKxERNV9MRpqw0T1bwcZajsupeYiOz5A6HCIiogbBZKQJc1RZY3QPQyHrxmgWshIRUfPEZKSJKytk/fV0MjLytRJHQ0REVP+YjDRx3Vs7o1srR2h1emw7xkJWIiJqfpiMWIDJYX4AWMhKRETNE5MRC/CPHj6wU8gRn56PI1duSx0OERFRvWIyYgHslVYY07MVAOBbFrISEVEzw2TEQpQVsv5+NhnpeRqJoyEiIqo/TEYsRFcfJwT7OqNYJ2LLURayEhFR88FkxIJMCTO0jmyKToBez0JWIiJqHpiMWJCRwd5wUFohIaMAf15OlzocIiKiesFkxILYKqwwrpehkHVjFAtZiYioeWAyYmEmhxvmHNkTl4LUnCKJoyEiIrp3TEYsTCcvB4T4uUCnF/H90USpwyEiIrpnTEYs0GRjIWsidCxkJSIiC8dkxAKN6O4NJxtr3MwqxMGLaVKHQ0REdE+YjFgglbUcj/RqDQD4loWsRERk4ZiMWKjJ4b4AgH3nU5CUXShxNERERHXHZMRCBXg6IKytK/QisDmGhaxERGS5mIxYsCml96v5LjoRJTq9xNEQERHVDZMRCza0mxdcbK2RnFOE/RdYyEpERJaJyYgFU1rJ8WhvQ+3IxqjrEkdDRERUN0xGLNyk0jlHDlxMw43MAomjISIiMh+TEQvX1t0Ofdu7QWQhKxERWag6JSNr1qyBv78/VCoVwsPDER0dXe32K1euRKdOnWBjYwNfX1/8+9//RlER76tSXyaXFbLGJKKYhaxERGRhzE5GNm/ejLlz52LhwoU4fvw4goODMWTIEKSmpla6/caNGzFv3jwsXLgQcXFx+Oyzz7B582b85z//uefgyWBwFy+42yuQlqtBZFyK1OEQERGZxexk5P3338fMmTMxY8YMdOnSBevWrYOtrS0+//zzSrf/66+/0K9fP0yePBn+/v4YPHgwJk2aVGNrCtWewkpmLGTljKxERGRpzEpGtFotjh07hoiIiDsHkMkQERGBI0eOVLpP3759cezYMWPycfXqVezatQvDhw+v8jwajQY5OTkmC1VvUqihq+bQpXQk3GYhKxERWQ6zkpH09HTodDqo1WqT9Wq1GsnJyZXuM3nyZCxZsgT3338/rK2t0b59ewwYMKDabprly5fDycnJuPj6+poTZovUxs0WD3RwBwBsimHrCBERWY4GH01z4MABLFu2DGvXrsXx48exfft2/PLLL1i6dGmV+8yfPx/Z2dnGJTGRo0Rqo2xG1i1HE6EtYSErERFZBitzNnZ3d4dcLkdKimmRZEpKCry8vCrd580338Tjjz+Op556CgAQFBSE/Px8PP3003j99dchk1XMh5RKJZRKpTmhEYCBgWp4OiiRmqvB7+eSMbK7j9QhERER1cislhGFQoGQkBBERkYa1+n1ekRGRqJPnz6V7lNQUFAh4ZDL5QAAURTNjZeqYS2XYWJo2Yys7KohIiLLYHY3zdy5c7F+/Xp8+eWXiIuLw7PPPov8/HzMmDEDADB16lTMnz/fuP2oUaPwv//9D9999x3i4+OxZ88evPnmmxg1apQxKaH6MzHUF4IA/HXlNuLT86UOh4iIqEZmddMAwMSJE5GWloYFCxYgOTkZPXr0wO7du41FrQkJCSYtIW+88QYEQcAbb7yBmzdvwsPDA6NGjcLbb79df1dBRq1dbDGgowf2X0jDpugE/Gd4oNQhERERVUsQLaCvJCcnB05OTsjOzoajo6PU4TR5e86lYOZXR+Fia40j8wdCZc0WKCIiany1/f7mvWmaoYc6ecDbSYXMgmL8drbyIddERERNBZORZsiqXCErZ2QlIqKmjslIMzUx1BcyAYiOz8Dl1FypwyEiIqoSk5FmytvJBg93NhQVb4zipHFERNR0MRlpxspmZN16LBFFxTqJoyEiIqock5Fm7MGOHmjlbIOcohL8cipJ6nCIiIgqxWSkGZPLBEwKK52RNZqFrERE1DQxGWnmJvT2hZVMwLHrmbiQzEJWIiJqepiMNHOejipEBJYVsl6XOBoiIqKKmIy0AJNLC1m3H7+JAm2JxNEQERGZYjLSAtwf4I42rrbI1ZTg55MsZCUioqaFyUgLIJMJmBRmaB35loWsRETUxDAZaSEe7d0a1nIBJxOzcPZWttThEBERGTEZaSHc7ZUY3NULALCR96shIqImhMlICzKltKtm54mbyNOwkJWIiJoGJiMtSJ/2bmjnbod8rQ4/xt6SOhwiIiIATEZaFEG4U8i6MZpzjhARUdPAZKSFeSSkNRRyGc7czMGpG1lSh0NERMRkpKVxtVNgWBALWYmIqOlgMtICTS7tqvkh9hZyiooljoaIiFo6JiMtUFhbVwR42qOwWIcfTtyUOhwiImrhmIy0QIIgGFtHvo1KgCiKEkdEREQtGZORFuqRXq2htJLhfHIuTiRmSR0OERG1YExGWignW2uM6O4NgIWsREQkLSYjLdiUcENXzU8nbyG7gIWsREQkDSYjLVivNi7o7OUATYke20/ckDocIiJqoZiMtGCCIGByaevIRhayEhGRRJiMtHBjeraCjbUcl1LzcPR6ptThEBFRC8RkpIVzVFljVDALWYmISDpMRgiTw/0AAL+cTkJmvlbiaIiIqKVhMkIIbu2Erj6O0Jbose04C1mJiKhxMRkh00LWaBayEhFR46pTMrJmzRr4+/tDpVIhPDwc0dHRVW47YMAACIJQYRkxYkSdg6b6N7pHK9gp5Lialo+/r2ZIHQ4REbUgZicjmzdvxty5c7Fw4UIcP34cwcHBGDJkCFJTUyvdfvv27UhKSjIuZ86cgVwux6OPPnrPwVP9sVda4R89WgEwtI4QERE1FrOTkffffx8zZ87EjBkz0KVLF6xbtw62trb4/PPPK93e1dUVXl5exmXPnj2wtbVlMtIElc3IuvtMEtLzNBJHQ0RELYVZyYhWq8WxY8cQERFx5wAyGSIiInDkyJFaHeOzzz7DY489Bjs7uyq30Wg0yMnJMVmo4XVr5YTg1k4o1onYeoyFrERE1DjMSkbS09Oh0+mgVqtN1qvVaiQnJ9e4f3R0NM6cOYOnnnqq2u2WL18OJycn4+Lr62tOmHQPygpZN0UnQK9nISsRETW8Rh1N89lnnyEoKAhhYWHVbjd//nxkZ2cbl8TExEaKkEYF+8BBaYXrtwvw15XbUodDREQtgFnJiLu7O+RyOVJSUkzWp6SkwMvLq9p98/Pz8d133+HJJ5+s8TxKpRKOjo4mCzUOW4UVxvQsK2S9LnE0RETUEpiVjCgUCoSEhCAyMtK4Tq/XIzIyEn369Kl23y1btkCj0eCf//xn3SKlRlPWVfP72RSk5hZJHA0RETV3ZnfTzJ07F+vXr8eXX36JuLg4PPvss8jPz8eMGTMAAFOnTsX8+fMr7PfZZ59hzJgxcHNzu/eoqUEFejuiVxtnlOhFbDnKQlYiImpYVubuMHHiRKSlpWHBggVITk5Gjx49sHv3bmNRa0JCAmQy0xznwoUL+PPPP/H777/XT9TU4CaH++F4QhY2RSfg2f7tIZMJUodERETNlCBawNzfOTk5cHJyQnZ2NutHGklRsQ5hb+9FTlEJNswIxYBOnlKHREREFqa239+8Nw1VSmUtx7herQEAG6M4IysRETUcJiNUpbIZWSPPpyI5m4WsRETUMJiMUJU6qB0Q5u8KnV7E5hjO9UJERA2DyQhVq2yY7+aYBOg4IysRETUAJiNUraHdvOBia41b2UU4cKHyOzMTERHdCyYjVC2VtRyPsJCViIgaEJMRqtGk0q6a/RdScTOrUOJoiIiouWEyQjVq72GPPu3coBeBzdFsHSEiovrFZIRqxVjIejQRJTq9xNEQEVFzwmSEamVIVy+42SmQkqNB5HkWshIRUf1hMkK1orCSYXxvFrISEVH9YzJCtTY5zNBVc/BSGhIzCiSOhoiImgsmI1Rrfm52eKCDO0QR2MRCViIiqidMRsgsZa0j3x+9gWIWshIRUT1gMkJmieiihoeDEul5Guw5lyJ1OERE1AwwGSGzWMtlmMBCViIiqkdMRshsj4W2gSAAf15Ox7X0fKnDISIiC8dkhMzm62qL/h09ALCQlYiI7h2TEaqTskLWLcduQFOikzgaIiKyZExGqE4e7uwJL0cVMvK1+O0sC1mJiKjumIxQnVjJZZgQ6gsA+Pbv6xJHQ0RElozJCNXZY6G+kAlAVHwGLqfmSR0OERFZKCYjVGc+zjZ4uLMnABayEhFR3TEZoXsyOdxQyLrt+A0UFbOQlYiIzMdkhO5J/46eaOVsg6yCYvx6JknqcIiIyAIxGaF7IpcJmGgsZGVXDRERmY/JCN2ziaG+kMsEHL2eiYspuVKHQ0REFobJCN0ztaMKEYGGQlber4aIiMzFZITqxeRwPwCGQtZCLQtZiYio9piMUL14IMAdvq42yC0qwc+nbkkdDhERWRAmI1QvZDIBj4Uahvl+y64aIiIyQ52SkTVr1sDf3x8qlQrh4eGIjo6udvusrCzMmjUL3t7eUCqV6NixI3bt2lWngKnpmtDbF1YyAbGJWTh3K0fqcIiIyEKYnYxs3rwZc+fOxcKFC3H8+HEEBwdjyJAhSE1NrXR7rVaLQYMG4dq1a9i6dSsuXLiA9evXo1WrVvccPDUtHg5KDOnqBQDYGM371RARUe0IoiiK5uwQHh6O0NBQrF69GgCg1+vh6+uLF154AfPmzauw/bp167BixQqcP38e1tbWdQoyJycHTk5OyM7OhqOjY52OQY3j8OV0TPk0CvZKK0T9ZyDslFZSh0RERBKp7fe3WS0jWq0Wx44dQ0RExJ0DyGSIiIjAkSNHKt3nxx9/RJ8+fTBr1iyo1Wp069YNy5Ytg05X9YgLjUaDnJwck4UsQ592bvB3s0WepgQ/nWQhKxER1cysZCQ9PR06nQ5qtdpkvVqtRnJycqX7XL16FVu3boVOp8OuXbvw5ptv4r333sNbb71V5XmWL18OJycn4+Lr62tOmCQhmUzApDAWshIRUe01+GgavV4PT09PfPLJJwgJCcHEiRPx+uuvY926dVXuM3/+fGRnZxuXxMTEhg6T6tH4kNZQyGU4fTMbp29kSx0OERE1cWYlI+7u7pDL5UhJSTFZn5KSAi8vr0r38fb2RseOHSGXy43rAgMDkZycDK1WW+k+SqUSjo6OJgtZDjd7JYZ2YyErERHVjlnJiEKhQEhICCIjI43r9Ho9IiMj0adPn0r36devHy5fvgy9Xm9cd/HiRXh7e0OhUNQxbGrqJocbump+iL2F3KJiiaMhIqKmzOxumrlz52L9+vX48ssvERcXh2effRb5+fmYMWMGAGDq1KmYP3++cftnn30WGRkZePHFF3Hx4kX88ssvWLZsGWbNmlV/V0FNTnhbV7T3sEOBVocfYlnISkREVTN73OXEiRORlpaGBQsWIDk5GT169MDu3buNRa0JCQmQye7kOL6+vvjtt9/w73//G927d0erVq3w4osv4rXXXqu/q6AmRxAMhaxv/RKHb6MSMCW8DQRBkDosIiJqgsyeZ0QKnGfEMmUVaBG2LBLaEj12PNcXPdu4SB0SERE1ogaZZ4TIHM62CowM8gYAbOQwXyIiqgKTEWpQZYWsP526hexCFrISEVFFTEaoQYX4uaCj2h5FxXrsPHFT6nCIiKgJYjJCDUoQBEw2zsh6HRZQokRERI2MyQg1uLG9WkNlLcPFlDwcu54pdThERNTEMBmhBudkY41R3X0AsJCViIgqYjJCjaKskPXn00nIKqj8NgBERNQyMRmhRtHD1xmB3o7Qluix7TgLWYmI6A4mI9QoBEHAlHAWshIRUUVMRqjRjO7hA1uFHFfT8hEVnyF1OERE1EQwGaFG46CyxugeLGQlIiJTTEaoUU0O8wMA7D6TjIx8FrISERGTEWpkQa2dENTKCVqdHluPJUodDhERNQFMRqjRlRWyboxKgF7PQlYiopaOyQg1ulHBPrBXWuHa7QIcuXpb6nCIiEhiTEao0dkprTCmJwtZiYjIgMkISaKskPW3s8lIy9VIHA0REUmJyQhJoouPI3r4OqNEL2ILC1mJiFo0JiMkmbJC1k3RLGQlImrJmIyQZEZ294GDygqJGYU4dDld6nCIiEgiTEZIMjYKOR7p1RoAsDHqusTREBGRVJiMkKQml3bV7I1LRUpOkcTREBGRFJiMkKQ6qh3Q288FOr2I72NYyEpE1BIxGSHJTbnvTiGrjoWsREQtDpMRktywbt5wtrXGrewi/HExVepwiIiokTEZIcmprMsXsnJGViKilobJCDUJk8IMXTX7zqfiVlahxNEQEVFjYjJCTUKApz3C27pCLwKbWchKRNSiMBmhJmPKfYb71XwXk4ASnV7iaIiIqLEwGaEmY0hXNVztFEjJ0WDfeRayEhG1FExGqMlQWsnxaEhpIWs0C1mJiFoKJiPUpJQVsv5xMQ2JGQUSR0NERI2hTsnImjVr4O/vD5VKhfDwcERHR1e57YYNGyAIgsmiUqnqHHC9KsoGSrRSR0Hl+LvboV+AG0QWshIRtRhmJyObN2/G3LlzsXDhQhw/fhzBwcEYMmQIUlOr7uN3dHREUlKScbl+vYncFO3PlcA7bYHvpgBHvwCyb0gdEQGYEm4oZN18NBHFLGQlImr2zE5G3n//fcycORMzZsxAly5dsG7dOtja2uLzzz+vch9BEODl5WVc1Gr1PQVdb24eA7R5wPmfgZ/nAB90Bdb2AX5/E4g/yFYTiQzqooa7vRJpuRrsPZcidThERNTAzEpGtFotjh07hoiIiDsHkMkQERGBI0eOVLlfXl4e/Pz84Ovri9GjR+Ps2bPVnkej0SAnJ8dkaRCP7wSe/gN46A3ANxwQZEDqOeCvj4AvRwHvtDO0mhzbAGTfbJgYqAJruQwTerOQlYiopTArGUlPT4dOp6vQsqFWq5GcnFzpPp06dcLnn3+OH374Ad988w30ej369u2LGzeq7hJZvnw5nJycjIuvr685YdaeTAb49AD6vwI8+TvwyhVg/OdA8CTAzgPQ5hpaTX56EfigC7C2L7BnARB/CNAVN0xMBMBQyCoIwKFL6bh+O1/qcIiIqAEJoijW+japt27dQqtWrfDXX3+hT58+xvWvvvoq/vjjD0RFRdV4jOLiYgQGBmLSpElYunRppdtoNBpoNBrj45ycHPj6+iI7OxuOjo61Dffe6PVA8kng0l7g0u/AzaOAWK5+QeEAtB8ABAwCOgwCHH0aJ64WZOrn0Th4MQ3P9G+PecM6Sx0OERGZKScnB05OTjV+f1uZc1B3d3fI5XKkpJj246ekpMDLy6tWx7C2tkbPnj1x+fLlKrdRKpVQKpXmhFb/ZDLAp6dh6f8KUJABXNkHXNoDXN4LFKQDcT8ZFgBQdwMCIgyJiW84ILeWNv5mYEp4Gxy8mIYtRxMxd1BHKKw4Ep2IqDky6393hUKBkJAQREZGGtfp9XpERkaatJRUR6fT4fTp0/D29jYvUqnZugJB44FxHwMvXwJm7gcG/AdoHQpAAFLOAIdXAhtGGGpNNj8OHP8KyLkldeQWa2BnT6gdlbidr8VvZyvvBiQiIstnVssIAMydOxfTpk1D7969ERYWhpUrVyI/Px8zZswAAEydOhWtWrXC8uXLAQBLlizBfffdh4CAAGRlZWHFihW4fv06nnrqqfq9ksYkkwGtehmWAa+VazX5vbTV5DYQ96NhAQB1ENAhwtCl4xvGVpNaspLLMLG3Lz7adxkboxIwKphdYUREzZHZycjEiRORlpaGBQsWIDk5GT169MDu3buNRa0JCQmQye40uGRmZmLmzJlITk6Gi4sLQkJC8Ndff6FLly71dxVSK2s1CRpvqDVJOmHozrm0xzB8OOW0YfnzA0DpdKfWJCACcLSwFqJGNjGsDVbvv4wjV2/jaloe2nnYSx0SERHVM7MKWKVS2wKYJin/9p1WkyuRhlaT8tRBhjqTDoOA1mGA3Oz8sNl7YkMM9p1PxcwH2uL1Ec0oiSUiauZq+/3NZKQx6XXArdjS7pw9wM3jAMr9+pVOQPuHDIlJQATgULui4OYuMi4FT355FM621vh7/kCorOVSh0RERLXAZMQS5KeXqzWJBAozTJ/36l6amAwyFMq20FYTnV7EA//dh1vZRVg5sQfG9GwldUhERFQLTEYsjV4H3DphSEwu7TH8XL7VROUEtH/4Tq2JQxOZUr+RfLj3Ej7YexFh/q74/pnajdwiIiJpMRmxdHlphhqTS3sM/xZmmj7v1R3oMNjQctKqd7NvNUnOLkK//+6DTi9iz78fRAe1g9QhERFRDZiMNCd6naG+pKzW5NYJ0+dVzoZWk7JaE3tPScJsaDO/Ooo951Iwo58/Fo7qKnU4RERUAyYjzZmx1aS01qQoy/R57x7lak16A7LmUfB54EIqpn8RA0eVFaJfj2AhKxFRE8dkpKXQ6wxzmZTVmiTFmj6vcgYCBpbWmgy06FYTvV7Egyv240ZmId59NBjjQ1pLHRIREVWDyUhLlZdqmAX20h7DSJ1KW03Kak1CLK7VZM3+y1jx2wX0auOM7c/1kzocIiKqBpMRAnQlhrsNX9pjqDVJOmn6vI0L0H6gITFpPxCw95AmTjOk5hah7/J9KNGL2D3nAXT24vuBiKipYjJCFeWm3Kk1ubIPKMou96RguENxWa1Jq15NttXk2W+O4dczyZjaxw9LRneTOhwiIqoCkxGqnrHVpLTWJPmU6fM2rqa1Jnbu0sRZiT8vpeOfn0XBQWmFqNcHwlbRvIc1ExFZKiYjZJ7c5HK1JvsBTWWtJqW1Jj49JW010etFPPTeAVy/XYD/PhKEiaFtJIuFiIiqxmSE6k5XAtyIvlNrknza9HkbV8N8JmW1JnZujR7iuj+u4P9+PY/g1k744fn7G/38RERUMyYjVH9ykgytJpf3AFcOVGw1aRVyp9bEpycgkzV4SOl5GvRZHolinYifX7gf3Vo5Nfg5iYgsXnERUJBuuIN8wW3DneXLfg56FPDoWK+nYzJCDUNXDNyIKa012Quk3NVqYutmaDUJGGSYFbYBW01e2HQCP528hcnhbbBsbFCDnYeIqEnS6wy3CjEmFuWSjPJLfjpQkGH4uTi/6uON/xzo9ki9hshkhBpHzq07tSZXDwCanHJPlrWaDAY6RADe9dtqcuTKbUxa/zdU1jI8FtoGAwM9Ed7WDQqrhm+ZISKqV6IIaPMqb7EwtmRkmCYchZkwuaFqbcmsDH842roDtq6Gn+3cgeBJhlm76xGTEWp8umIgMaq01mQvkHLG9Hlb93K1Jg8bPgT3QBRFTPzkb0THZxjX2Sut8GBHd0QEqvFQJ0+42Cnu6RxERHVSogUKM2rfYlFwG9Bp6nYulVNpYuFWmli43fm5svVKR0AQ6vd6q8BkhKSXfdO01kSbe+c5QXan1SQgwjAzbB1aTYqKdfjzUjr2xqUg8nwq0nLvfJhlAhDi54KBgWpEBKrR3sMOQiN9AImoGdHrDbVy+XcnFFW0WBTcvquV2AxWqootFrZuFZey9TYugNy6fq+3HjEZoaalRGtoNbm8x1BrknrW9Hk7j9Jak4g6t5ro9SJO38zG3rgU7I1LRVyS6X8G/m62GBioxsBAT4T6u8Jazu4cohapuPCu5CHjrqLO8i0WpT+LOvPPI8gMow+NyYNr9S0Wtm6Awq7+r1dCTEaoacu+WZqY7AGu/lFJq0nvO7UmXsF1ajW5mVWIfXEp2BOXir+v3IZWpzc+56iywoBOnhgY6IkBnTzhZNN0/7IgomrodaZdHTW1WBTcBooL6nYuhYMhoaiqteLulgyVc6OMLmzKmIyQ5SjRAol/36k1ST1n+rydZ2mtSWmriY2L2afI05Tgz0tp2BuXin3nU5GRrzU+ZyUTEOrvioGBnogIVMPfvXn9ZUJkMUQR0OSWa624XUOLxW2gMAt1K+K0rqHFopKkw0pZ31fc7DEZIcuVfeNOYnL1gKHCvIwgA1qH3pnXxKu72X956PQiYhMzsTcuFZFxKbiYkmfyfHsPO0R0MdSZ9GrjArmMdSZEdVKiraTFooaiTp225uNWRuVcyxYLV0PCoXRotCLOlozJCDUPJVog4cidWpO0ONPn7TyBNveVK+ZyNfxnU9ZPa+ti+FnlVOV/PAm3C0oLYFMQdTUDJfo7HwkXW2s81NnQYvJAB3c4qNidUx/0oh4XMy8iJjkGx1OOI1ebC5kgg0wmgwwyyAU5ZIIMcpnhXxkMzxnXC3IIgmDyWCbIjItZ25UeV4BgPF9djlftz9Wco7JjNslC6xKNIZEozKj83wotGRn3UMRpc1eLRVUjRUqfs3EB5LxHVVPEZISap6zEO4nJ1QPVT+BTnszK8B+WMUlxNTwu+8+uNInJt3JCdDKw91oxdl0uQmbRnToTa7mA+9q5YWBnTwwMVMPX1bZhrrEZ0ot6XMq8hKMpRxGdFI2jKUeRo63jF1ULIECoXSJUmryVT9xqTMAggwwi5HodBL0Ocn0JZLqS0n+LSxct5DotZCVayEs0EEo0kOuLIQMgFwEZRMgAyERADrH0X0CAWPq8YRu5CMgEATJrW8gUdpBb20OmsIdcaQ9BYQ+50hEypYNhUTlBrnSCTOUEmbVNrZK/ypK4qhLJmpLGJpkANgNMRqj5K9EYWk1Sz5f76+x2ub/YMu+pWE2EAJ3CETmCA5KLbZFUbIcs2CNTtEeG6ACVozv8fFujS/u2aN+mDWR2pUkO+5UhiiIuZ11GTHIMYpJjcDTlKLI0WSbb2FjZoJe6F0LVofC284ZO1EEv6qEX9VX+XP5xpev11W8vimL1+1e1nV4PPSqeR4QInV5n9nmo6REg3FOCU10CaLLurta/u1vQyieXZTFVeFxJEirDXQlpuVa4u1sWq4q7q1tXeNl51evvlckIUZniojuTDxmblW8DBZmVJzEFGXfdf8c8osIOgk1pF1H5riOTbiSXci00robhfBb8l5koiriafRUxyTGITo7GsZRjyCjKMNnGxsoGPT17ItQrFKFeoeji1gXWspbZ7VUhadHroCvMhL7wNvQFmdAXZkBXkAF9USZ0hZkQi7KgK8yCvigLuqJs6DU50BdlQ19SBJ0A6AHoIEBf2c+CAD1Mf9YBEK2U0CnsoVfYQqewg97aFnqFLfTWNtBZ20BvrYLOSgXR2gY6uQJ6KyV0cmvoIVZMuPSliVllj8sncnpdhaSu/HPl9zEme/ewnV6885hq9t8H/ovh7YbX6zFr+/3NTjZq/qxVgLUP4OhT+310JYaWlUqTmAxoctORkZaEgqw0CIWZcEQuXJALuSBC0OYD2nwgO6H255Mry9W6uN71890JTWn3ktJJsmGDoigiPiceR5OPIjo5GjHJMRWSD5VcZZJ8dHXv2jKSD+PMm1XVV2RCVpgBmcn6zLrNYwGUzmXhcle9VLluyArrS5+zVtXvdTdhoiiaJCYmiUolrVzl11dooavkucpaxMxp5avqGHU9Zm2Pcfd2LirzRyrWFyYjRJWRWwH2HoalEkoA3qU/a0v0iLmWgb3nkvB3XDwKstLgilw4C3lwQS4CnUvQ3VWHAHstXGV5EAoyTJMbndYwDXRukmGpLUF+1xdOuYLd8q0u5X+uY6GfKIq4nnMdMSkxiEmKQUxKDNIL001/J3Ilenj2QKg6FGHeYejm1g3WTXhmyBqJIlCUbUwgTBKHKpIMFGaYjv4yl7Vt6WtWWXJRRbIhYVJqKcq6IuSQG1bIpY2HKmI3DVE9EkURl1LzDKNz4lJxPCET5T9hakclHu6sxqAunujb3h0qK5mhFcWkuyizkvqXu7qW7uULT+VURavLnS86UeWCRJQgJv86ojPicDT1BFILU00Oo5Ap0MOzB3p79UaYVxiC3IOgkDfRewHVorWi4voGaq2orsWiBbVWUMvAmhGiJiA9T4P951MRGZeKg5fSUKC98+Wmspbh/gAPRAR64uFAT3g6mPFFZDLM8jYqDLGsrBamKKvKw4kAbljJcVSlQrSNCjEqJVKsTFtQrEUR3fVWCJM7IFSlRnc7XyjtPMolNHfVyCjs678ORhQNw0WrTSLYWkHUVDAZIWpiiop1+PvqbUSWTrZ2K7vI5PlgX2dEdPZERBc1Ons51P9QQ12JISEp/aK+mXkJMemnEZN9GTEFN5CkKzTZ3EoU0V2jRWhhEUKLihCs0UJlzn8XckXFL+vKuo5sXQ13fK5Ni0VhJqAvqdv1CzLDxFjVJRFsrSCqV0xGiJowURQRl5Rb2p2TgpM3TEfvtHK2wcBAw3wm97VzhdLq3ju5k/KSEJMSY5zn42beTZPnrQQrdHPvZiw47eHZAzZylWndRKVdR5W0ypQUVRFFPamqtaLKpMKF9wkhkkCDJiNr1qzBihUrkJycjODgYKxatQphYWE17vfdd99h0qRJGD16NHbu3Fnr8zEZoeYuNacI+86nYm9cCv68nI6i4jtDEe0UcjzY0QMDA9V4qJMH3OxrN49Jcn6ycZ6PmOQY3Mi7YfK8XJCjq3tXhHmFIVRtSD5sretpIjdtwV3DpWuohamsFaW6xMLapn7iJKIG1WDJyObNmzF16lSsW7cO4eHhWLlyJbZs2YILFy7A09Ozyv2uXbuG+++/H+3atYOrqyuTEaIqFGp1+OtKurEINjVXY3xOEIBebVwQEahGRKAnAjztjd05qQWpJslHQq7p0GK5IEdXt67GgtOenj3rL/kgIqpEgyUj4eHhCA0NxerVqwEAer0evr6+eOGFFzBv3rxK99HpdHjwwQfxxBNP4NChQ8jKymIyQlQLer2IM7eyjTf1O3vrzhTqglUO1B434am+gQLhIlKKEk32lQkydHHtYux26enZE/YK+8a+BCJqwRpk0jOtVotjx45h/vz5xnUymQwRERE4cuRIlfstWbIEnp6eePLJJ3Ho0KEaz6PRaKDR3PlrMCeH97CglkkmE9C9tTO6t3bG1PtdsefqX9h16U+cz4pFkZCEfADxZR8VUYC9zA/BbiH4R+cH8IBvGBwUDlKGT0RUK2YlI+np6dDpdFCr1Sbr1Wo1zp8/X+k+f/75Jz777DPExsbW+jzLly/H4sWLzQmNqNnJKMowznB6NPkormRfufOkYLiXho9Ne1gVB+BmUitkZrRGrt4GSQD2HNagt98ZRASqMTDQE+082CJCRE1Xg87Ampubi8cffxzr16+Hu7t7rfebP38+5s6da3yck5MDX1/feo8vJT8FAOBu4w65jFPykbQyizJxLOWYcXr1y1mXK2zTyaWTsdslRB0CJ6UTAECnFxGbmIXI0jqTCym5iIrPQFR8Bt7eFYd2HnaGxKSzJ0L8XGAl56gSImo6zEpG3N3dIZfLkZKSYrI+JSUFXl4V7/R35coVXLt2DaNGjTKu0+sNowSsrKxw4cIFtG/fvsJ+SqUSSmXD3/l07cm12H5pO6wEK3jYesDLzgtetl7wsvOC2k5teFy6zlXlyltMU73K1mTjaMpRY8HpxcyLFbbp4NLBML26VxhC1CFwVjlXeiy5TECInwtC/Fzw6tDOSMwoMBbARsXfxtW0fHySdhWfHLwKZ1trPNTJEwMDPfFgRw84qix4ynYiahbMSkYUCgVCQkIQGRmJMWPGADAkF5GRkXj++ecrbN+5c2ecPn3aZN0bb7yB3NxcfPjhhw3S2mEOjU4DuSBHiViCpPwkJOVXfV8QhUxxJ0EpTVjKFrWtYb2jwpEJC1UpW5ON4ynHDd0uKUdxIeMCRJjWjwc4B6C3ujfCvA3Jh6vKtU7n8nW1xYx+bTGjX1vkFhXj4EXD6Jz9F1KRVVCMHSduYseJm7CSCQhv51o6OkcNX1eOriGixlenob3Tpk3Dxx9/jLCwMKxcuRLff/89zp8/D7VajalTp6JVq1ZYvnx5pftPnz69SY2m0el1SCtMQ0pBCpLzk41L+cfphekVvjQqY2NlU2myUv4xh1K2HLnaXGPyEZMcg/MZ5yu8j9o5tTN2u/RW94abjVuDxlSi0+N4gqE7Z09cCq6m5Zs831Ftj4GliUkPX2fIZUyuiajuGmQ0DQBMnDgRaWlpWLBgAZKTk9GjRw/s3r3bWNSakJAAmQXNciiXyY2JQrBHcKXbFOuKkVqYapKsJOcnI7kgGSn5hqQlU5OJwpJCxGfHIz47vsrzOSgcqk1Y1HZqKOUN30VF9S9Pm4fjqceN3S5xGXHQi3qTbfwd/RHqZeh26e3VG+42ta+lqg9WchnC2roirK0r5g8PRHx6PiLjUrA3LgUx1zJxMSUPF1Py8L8DV+Bmp8BDnT0REeiJBzp4wE7Jm3wTUcPgdPD1pKikqELrSnJBuZaW/BTkFufW6liuKldj109lrSseth6wlrGfX2r5xfk4kXrCONrl3O1z0N11l1c/Rz9Dt0tp8uFpW/XEgFLLLijGgYup2BuXigMXUpFbdOceMAq5DH3auyGidIp6H2fOgEpENeO9aZqg/OL8KpOVsq6hwpLCGo8jQICHjYexJaWyxIUjhOpfQXEBYlNjDd0uKTE4m362QvLh6+Br0u3iZVexsNsSFOv0iInPMEy2dj4F128XmDzfxdvRmJgEtXKCjN05RFQJJiMWSBRF5GhzakxYivXFNR6LI4TuXWFJIWJTY43dLmfSz6BENL1jbCv7VsZul1CvUItNPqojiiIup+YZZ4E9npAJfbn/NTwdlIab+nVWo1+AO2wUTIKJyIDJSDOlF/XIKMow1qrcnawkFyQjrSCtwl/sleEIIVNFJUU4mXbS2O1yKv0USu66Xb23nbdJ8uFj7yNRtNK5nafB/gtpiIxLwcGLacjX3nmvKa1kuD/AHRFdDHOaeDqqJIyUiKTGZKQFK9GXIL0wvUKRbfmEJb0wvVbHas4jhDQ6DU6lnTKOdjmVdqpCq5PaVm1MPEK9QtHaobVE0TZNmhIdoq5mlBbBpuJmlmk3Y/fWThjY2TALbFeflpPYEpEBkxGqVrGu+E7B7V2FtmWPszRZtTqWpYwQ0uq0OJV2ytDtkhKDk6knodVrTbbxtPFEqHdpy4fakHzwC7R2RFHE+eTc0mHDqTiZmGXyvI+TCg+X1pn0aecGlTW7c4iaOyYjdM8KSwpNkpPKhjTnFefV6lhSjBAq1hXjdPppY7dLbFosNDqNyTYeNh7o7dXb2PrRxqENk496kppbhP3nDaNz/ryUjsLiO905tgo5HujgjoGBajzc2RPu9tInq0RU/5iMUKPI0+ZVWbtSlrAU6YpqPI5MkMFd5W5aaHtXS4u7jTtkQtVz2BTrinH29lljt0tsamyFc7up3IxdLqFeofB39Gfy0QiKinX460q6sQg2JedOUigIgI+TDdwdlHC3U8DNXgE3eyXc7BRwt1caHtsp4W6vgIudAta8rw6RxWAyQk2CKIrI1mRXmqyUHyF0d6FoZawEK3jaelZIWApKChCTHIMTqScqDI12Vbka5/kI9QpFW6e2TD4kJooiztzMMdw753wKztzMMWt/Z1truNkZEhb30kSlLIFxL13vZq+Au50SjjZWfL2JJMRkhCxG2Qihyma4Lfs5rTCtwmymlXFWOt9p+VCHor1ze34ZNXGpuUW4kVmI23la3M7T4Ha+Ful5GsPjfMO/6XlaZORrTIYU14a1XICr3Z2Exb20xcWYsNibPsc6FqL6xWSEmpXqRggJgoAQdQhCvUIR4BxQbVcOWS69XkRWYTFu52mQXi5RuZ2nQXp+aSKTpzUmM+VnkK0te6VVabdQzS0vLrYK3ruHqAZMRoioRdOU6EqTFS3SyyUulbW83M7TQqurueWtPEEAXG0VxpqWmlpe7JXsMqKWp8FulEdEZAmUVnL4ONvU6j46oigiV1Nyp6WlhpaXzAItRBG4nW94DNQ8qkxhJTNpWSkryq2QzNgr4GqngNKKXUbUcjAZIaIWTxAEOKqs4aiyRlt3uxq3L9HpkVlQXK6mpWKNS/lkJl+rg7ZEj1vZRbiVXfPoMgBwVFmZjCYydhWZJC+Gn51srHl/ILJoTEaIiMxkJZfBw0EJD4fazY9SqNUZEpZyLSwVu44MP2fka1GiF5FTVIKcohJcTc+v8fhyWVmhrqJCAnN3ka6bvQK2Cv7XT00L35FERA3MRiGHr6stfF1rvnWCXi8ip6jYmJyUJTDpd9W4lCUz2YXF0OlFpOVqkJarAZBbczzW8ruKcjm3C0mLyQgRURMikwlwtlXA2VaBAE/7GrfXluiRWVBFV9FdyUx6ngaaEj0Ki3W4kVmIG5mFNR4fMJ3bxaOylhd7JbwcVVA7qqCwYuJC5mMyQkRkwRRWMqhLE4GaiKKIfK3uTktLhZaX8oW6hi4jvQhkFRQjq6AYV9Jq7jJyt1fCx1kFbycVvJ1sDP8628Cn9F9PByVbWqgCJiNERC2EIAiwV1rBXmkFP7eaC3V1ehFZBVrT4dB31bjcztciLVeD5JwiaEv0SM/TID1Pg1M3sis9pkwAPByU8HaygY+zCl6ONqXJiw28S5MYTwcV53BpYZiMEBFRpeQyoXQoshId1Q7VbiuKIjLytUjKLsKtrEIk5xThVlYRkrILkZRVhKScQiRnF6FYJyIlR4OUHA1iE6s+r9pBCW9nG3g5qQytKmXJi5OhlcXdXskRRM0IkxEiIrpngnAncenWyqnSbfR6Een5GkNykl2aqJQlL9mGdck5RdDpxRqHQVvLBagdy3UHOavg41SWvBgeu9kpONGchWAyQkREjUImE+DpYOiGCfatfJuykUHlE5Wk7CIkZxfhVmkrS2quoYXlThFuZqXHUshl8HIyJCw+zqX1K3clL8621kxYmgAmI0RE1GTIZQK8nFTwclKhZxXblOj0SC1NWIxdQdlFpS0uhbiVXYT0PA20Oj0SMgqQkFFQ5flU1rI7hbbGgts7rSvejja8+3MjYDJCREQWxUouM071H+JX+TbaEj1Scky7g5KyDIlKUrahWyg9T4uiYj3i0/MRX83kcrYKubF1xcvRdHRQWWuLg8q6ga62ZWAyQkREzY7CSlbjRHNFxTqklBbaJucUmhbcliYtmQXFKNDqcCUtv9qhzQ5KK3iXK7AtPzqorPiWM99Wjb8ZIiJqkVTWcvi52VU7zLlQqzO2pNwq17qSXK6mJaeoBLmaEuSm5OFiStU3TXSysb5Tt1LaumJMXkpbWVTWLfMGiUxGiIiIqmCjkKOdhz3aeVQ9G26+puROd1CWodC2fPKSlF2EPE0JsguLkV1YjPPJVU/Z72JrbWxJ8S4bHVT6s4+TDdROymZ5R2cmI0RERPfATmmFAE/7aqfvzy0qNhkdlFQuUSkbJVRYrENmQTEyC4pxLimnymO52yvKFd3eaVUpq2nxclJZ3Cy3TEaIiIgamIPKGg4q6yonjxNFETmFJYbEpNzooLJExTCJXCE0JfrS+wxpcfpm5bPcCgLgYW+YNM7b8a7RQaU1LJ4OSlg1oYSFyQgREZHEBEGAk601nGytEejtWOk2oigis6C43CRxhSZdQWXzsWhLhz6n5mpwsorzyQRAXdqK4lPayvKPHj7o3tq5wa6xOkxGiIiILIAgCHC1U8DVTlHtLLe387XlJokr1y1UOi9LSk4RSvSicf0JZAEAglo7MRkhIiKieyOTCfBwUMLDQYmg1pUnLDq9iNt5mrtaVQrR1afyFpnGUKcOozVr1sDf3x8qlQrh4eGIjo6uctvt27ejd+/ecHZ2hp2dHXr06IGvv/66zgETERFR3cllAjwdVejh64xhQd544v62eH1EFwR4Vn8zxIZkdjKyefNmzJ07FwsXLsTx48cRHByMIUOGIDU1tdLtXV1d8frrr+PIkSM4deoUZsyYgRkzZuC333675+CJiIjI8gmiKIrm7BAeHo7Q0FCsXr0aAKDX6+Hr64sXXngB8+bNq9UxevXqhREjRmDp0qW12j4nJwdOTk7Izs6Go6N0zUhERERUe7X9/jarZUSr1eLYsWOIiIi4cwCZDBEREThy5EiN+4uiiMjISFy4cAEPPvhgldtpNBrk5OSYLERERNQ8mZWMpKenQ6fTQa1Wm6xXq9VITk6ucr/s7GzY29tDoVBgxIgRWLVqFQYNGlTl9suXL4eTk5Nx8fWt4l7TREREZPEaZcYTBwcHxMbGIiYmBm+//Tbmzp2LAwcOVLn9/PnzkZ2dbVwSExMbI0wiIiKSgFlDe93d3SGXy5GSkmKyPiUlBV5eXlXuJ5PJEBAQAADo0aMH4uLisHz5cgwYMKDS7ZVKJZRKpTmhERERkYUyq2VEoVAgJCQEkZGRxnV6vR6RkZHo06dPrY+j1+uh0WjMOTURERE1U2ZPejZ37lxMmzYNvXv3RlhYGFauXIn8/HzMmDEDADB16lS0atUKy5cvB2Co/+jduzfat28PjUaDXbt24euvv8b//ve/+r0SIiIiskhmJyMTJ05EWloaFixYgOTkZPTo0QO7d+82FrUmJCRAJrvT4JKfn4/nnnsON27cgI2NDTp37oxvvvkGEydOrL+rICIiIotl9jwjUuA8I0RERJanQeYZISIiIqpvTEaIiIhIUkxGiIiISFJmF7BKoayshdPCExERWY6y7+2aylMtIhnJzc0FAE4LT0REZIFyc3Ph5ORU5fMWMZpGr9fj1q1bcHBwgCAI9XbcnJwc+Pr6IjExsdmO0mnu18jrs3zN/Rqb+/UBzf8aeX11J4oicnNz4ePjYzLtx90somVEJpOhdevWDXZ8R0fHZvkGK6+5XyOvz/I192ts7tcHNP9r5PXVTXUtImVYwEpERESSYjJCREREkmrRyYhSqcTChQub9R2Cm/s18vosX3O/xuZ+fUDzv0ZeX8OziAJWIiIiar5adMsIERERSY/JCBEREUmKyQgRERFJiskIERERSYrJCBEREUmq2Scja9asgb+/P1QqFcLDwxEdHV3t9lu2bEHnzp2hUqkQFBSEXbt2NVKkdWfONW7YsAGCIJgsKpWqEaM1z8GDBzFq1Cj4+PhAEATs3Lmzxn0OHDiAXr16QalUIiAgABs2bGjwOOvK3Os7cOBAhddPEAQkJyc3TsBmWr58OUJDQ+Hg4ABPT0+MGTMGFy5cqHE/S/kc1uX6LO0z+L///Q/du3c3zs7Zp08f/Prrr9XuYymvH2D+9Vna63e3//u//4MgCJgzZ0612zX2a9isk5HNmzdj7ty5WLhwIY4fP47g4GAMGTIEqamplW7/119/YdKkSXjyySdx4sQJjBkzBmPGjMGZM2caOfLaM/caAcOUv0lJScbl+vXrjRixefLz8xEcHIw1a9bUavv4+HiMGDECDz30EGJjYzFnzhw89dRT+O233xo40rox9/rKXLhwweQ19PT0bKAI780ff/yBWbNm4e+//8aePXtQXFyMwYMHIz8/v8p9LOlzWJfrAyzrM9i6dWv83//9H44dO4ajR4/i4YcfxujRo3H27NlKt7ek1w8w//oAy3r9youJicHHH3+M7t27V7udJK+h2IyFhYWJs2bNMj7W6XSij4+PuHz58kq3nzBhgjhixAiTdeHh4eK//vWvBo3zXph7jV988YXo5OTUSNHVLwDijh07qt3m1VdfFbt27WqybuLEieKQIUMaMLL6UZvr279/vwhAzMzMbJSY6ltqaqoIQPzjjz+q3MYSP4dlanN9lvwZLOPi4iJ++umnlT5nya9fmequz1Jfv9zcXLFDhw7inj17xP79+4svvvhildtK8Ro225YRrVaLY8eOISIiwrhOJpMhIiICR44cqXSfI0eOmGwPAEOGDKlye6nV5RoBIC8vD35+fvD19a3xLwBLY2mvYV316NED3t7eGDRoEA4fPix1OLWWnZ0NAHB1da1yG0t+DWtzfYDlfgZ1Oh2+++475Ofno0+fPpVuY8mvX22uD7DM12/WrFkYMWJEhdemMlK8hs02GUlPT4dOp4NarTZZr1arq+xfT05ONmt7qdXlGjt16oTPP/8cP/zwA7755hvo9Xr07dsXN27caIyQG1xVr2FOTg4KCwsliqr+eHt7Y926ddi2bRu2bdsGX19fDBgwAMePH5c6tBrp9XrMmTMH/fr1Q7du3arcztI+h2Vqe32W+Bk8ffo07O3toVQq8cwzz2DHjh3o0qVLpdta4utnzvVZ4uv33Xff4fjx41i+fHmttpfiNbRqsCNTk9SnTx+TjL9v374IDAzExx9/jKVLl0oYGdVGp06d0KlTJ+Pjvn374sqVK/jggw/w9ddfSxhZzWbNmoUzZ87gzz//lDqUBlHb67PEz2CnTp0QGxuL7OxsbN26FdOmTcMff/xR5Re2pTHn+izt9UtMTMSLL76IPXv2NOlC22abjLi7u0MulyMlJcVkfUpKCry8vCrdx8vLy6ztpVaXa7ybtbU1evbsicuXLzdEiI2uqtfQ0dERNjY2EkXVsMLCwpr8F/zzzz+Pn3/+GQcPHkTr1q2r3dbSPoeAedd3N0v4DCoUCgQEBAAAQkJCEBMTgw8//BAff/xxhW0t8fUz5/ru1tRfv2PHjiE1NRW9evUyrtPpdDh48CBWr14NjUYDuVxuso8Ur2Gz7aZRKBQICQlBZGSkcZ1er0dkZGSVfYF9+vQx2R4A9uzZU23foZTqco130+l0OH36NLy9vRsqzEZlaa9hfYiNjW2yr58oinj++eexY8cO7Nu3D23btq1xH0t6DetyfXezxM+gXq+HRqOp9DlLev2qUt313a2pv34DBw7E6dOnERsba1x69+6NKVOmIDY2tkIiAkj0GjZYaWwT8N1334lKpVLcsGGDeO7cOfHpp58WnZ2dxeTkZFEURfHxxx8X582bZ9z+8OHDopWVlfjuu++KcXFx4sKFC0Vra2vx9OnTUl1Cjcy9xsWLF4u//fabeOXKFfHYsWPiY489JqpUKvHs2bNSXUK1cnNzxRMnTognTpwQAYjvv/++eOLECfH69euiKIrivHnzxMcff9y4/dWrV0VbW1vxlVdeEePi4sQ1a9aIcrlc3L17t1SXUC1zr++DDz4Qd+7cKV66dEk8ffq0+OKLL4oymUzcu3evVJdQrWeffVZ0cnISDxw4ICYlJRmXgoIC4zaW/Dmsy/VZ2mdw3rx54h9//CHGx8eLp06dEufNmycKgiD+/vvvoiha9usniuZfn6W9fpW5ezRNU3gNm3UyIoqiuGrVKrFNmzaiQqEQw8LCxL///tv4XP/+/cVp06aZbP/999+LHTt2FBUKhdi1a1fxl19+aeSIzWfONc6ZM8e4rVqtFocPHy4eP35cgqhrp2wo691L2TVNmzZN7N+/f4V9evToISoUCrFdu3biF1980ehx15a51/ff//5XbN++vahSqURXV1dxwIAB4r59+6QJvhYquzYAJq+JJX8O63J9lvYZfOKJJ0Q/Pz9RoVCIHh4e4sCBA41f1KJo2a+fKJp/fZb2+lXm7mSkKbyGgiiKYsO1uxARERFVr9nWjBAREZFlYDJCREREkmIyQkRERJJiMkJERESSYjJCREREkmIyQkRERJJiMkJERESSYjJCREREkmIyQkRERJJiMkJERESSYjJCREREkvp/AMoJhXHWfa8AAAAASUVORK5CYII=","text/plain":["<Figure size 640x480 with 1 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":["plt.plot(train_losses, label='train_losses')\n","plt.plot(valid_losses, label='valid_losses')\n","\n","# Move the test_losses tensor to the CPU before plotting\n","# Convert the test_losses list to a NumPy array before plotting\n","plt.plot(np.array(test_losses), label='test_losses')\n","\n","plt.legend(frameon=True)\n","plt.title('Training, Validation, and Test Losses')\n","plt.show()"]},{"cell_type":"code","execution_count":17,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:40:14.978030Z","iopub.status.busy":"2023-12-30T11:40:14.977535Z","iopub.status.idle":"2023-12-30T11:40:15.266850Z","shell.execute_reply":"2023-12-30T11:40:15.265647Z","shell.execute_reply.started":"2023-12-30T11:40:14.977986Z"},"trusted":true},"outputs":[{"data":{"text/plain":["Text(0.5, 1.0, 'Validation and Test Accuracy')"]},"execution_count":17,"metadata":{},"output_type":"execute_result"},{"data":{"image/png":"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","text/plain":["<Figure size 640x480 with 1 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":["plt.plot(val_acc, label = 'val_accuracy')\n","plt.plot(test_acc, label = 'test_accuracy')\n","plt.legend(frameon= True)\n","plt.title('Validation and Test Accuracy')\n"]},{"cell_type":"markdown","metadata":{},"source":["## Save the checkpoint\n","\n","Now that your network is trained, save the model so you can load it later for making predictions. You probably want to save other things such as the mapping of classes to indices which you get from one of the image datasets: `image_datasets['train'].class_to_idx`. You can attach this to the model as an attribute which makes inference easier later on.\n","\n","```model.class_to_idx = image_datasets['train'].class_to_idx```\n","\n","Remember that you'll want to completely rebuild the model later so you can use it for inference. Make sure to include any information you need in the checkpoint. If you want to load the model and keep training, you'll want to save the number of epochs as well as the optimizer state, `optimizer.state_dict`. You'll likely want to use this trained model in the next part of the project, so best to save it now."]},{"cell_type":"code","execution_count":19,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:51:27.555854Z","iopub.status.busy":"2023-12-30T11:51:27.555140Z","iopub.status.idle":"2023-12-30T11:51:27.577247Z","shell.execute_reply":"2023-12-30T11:51:27.575917Z","shell.execute_reply.started":"2023-12-30T11:51:27.555813Z"},"trusted":true},"outputs":[],"source":["# TODO: Save the checkpoint \n","\n","checkpoint = {'input_size': 224,\n","              'output_size': 102,\n","              'model_name': model_name,\n","              'state_dict': trained_model.state_dict(),\n","              'optimizer_state_dict': optimizer.state_dict(),\n","              'scheduler_state_dict': scheduler.state_dict(),\n","              'class_to_idx': class_to_idx\n","             }\n","\n","path_to_save = '/Users/dalex/Udacity/new_try/aipnd-project/'\n","\n","torch.save(checkpoint, path_to_save + 'checkpoint_5_ep_densenet_121_experimentation.pth')"]},{"cell_type":"markdown","metadata":{},"source":["## Loading the checkpoint\n","\n","At this point it's good to write a function that can load a checkpoint and rebuild the model. That way you can come back to this project and keep working on it without having to retrain the network."]},{"cell_type":"code","execution_count":30,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:51:45.519941Z","iopub.status.busy":"2023-12-30T11:51:45.518892Z","iopub.status.idle":"2023-12-30T11:51:45.526498Z","shell.execute_reply":"2023-12-30T11:51:45.525324Z","shell.execute_reply.started":"2023-12-30T11:51:45.519890Z"},"trusted":true},"outputs":[],"source":["# TODO: Write a function that loads a checkpoint and rebuilds the model\n","\n","def load_checkpoint(filepath):\n","    checkpoint = torch.load(filepath)\n","    name = checkpoint['model_name']\n","    model = build_model(name)\n","    \n","    optimizer.load_state_dict(checkpoint['optimizer_state_dict'])\n","    scheduler.load_state_dict(checkpoint['scheduler_state_dict'])\n","    model.load_state_dict(checkpoint['state_dict'])\n","    model.class_to_idx = checkpoint['class_to_idx']\n","    \n","    return model\n"]},{"cell_type":"code","execution_count":31,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:51:48.100484Z","iopub.status.busy":"2023-12-30T11:51:48.099771Z","iopub.status.idle":"2023-12-30T11:51:48.830522Z","shell.execute_reply":"2023-12-30T11:51:48.829636Z","shell.execute_reply.started":"2023-12-30T11:51:48.100440Z"},"trusted":true},"outputs":[{"name":"stderr","output_type":"stream","text":["/Users/dalex/miniforge3/envs/mps/lib/python3.10/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead.\n","  warnings.warn(\n","/Users/dalex/miniforge3/envs/mps/lib/python3.10/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing `weights=DenseNet121_Weights.IMAGENET1K_V1`. You can also use `weights=DenseNet121_Weights.DEFAULT` to get the most up-to-date weights.\n","  warnings.warn(msg)\n"]}],"source":["loaded_model = load_checkpoint('checkpoint_5_ep_densenet_121_experimentation.pth')"]},{"cell_type":"markdown","metadata":{},"source":["# Inference for classification\n","\n","Now you'll write a function to use a trained network for inference. That is, you'll pass an image into the network and predict the class of the flower in the image. Write a function called `predict` that takes an image and a model, then returns the top $K$ most likely classes along with the probabilities. It should look like \n","\n","```python\n","probs, classes = predict(image_path, model)\n","print(probs)\n","print(classes)\n","> [ 0.01558163  0.01541934  0.01452626  0.01443549  0.01407339]\n","> ['70', '3', '45', '62', '55']\n","```\n","\n","First you'll need to handle processing the input image such that it can be used in your network. \n","\n","## Image Preprocessing\n","\n","You'll want to use `PIL` to load the image ([documentation](https://pillow.readthedocs.io/en/latest/reference/Image.html)). It's best to write a function that preprocesses the image so it can be used as input for the model. This function should process the images in the same manner used for training. \n","\n","First, resize the images where the shortest side is 256 pixels, keeping the aspect ratio. This can be done with the [`thumbnail`](http://pillow.readthedocs.io/en/3.1.x/reference/Image.html#PIL.Image.Image.thumbnail) or [`resize`](http://pillow.readthedocs.io/en/3.1.x/reference/Image.html#PIL.Image.Image.thumbnail) methods. Then you'll need to crop out the center 224x224 portion of the image.\n","\n","Color channels of images are typically encoded as integers 0-255, but the model expected floats 0-1. You'll need to convert the values. It's easiest with a Numpy array, which you can get from a PIL image like so `np_image = np.array(pil_image)`.\n","\n","As before, the network expects the images to be normalized in a specific way. For the means, it's `[0.485, 0.456, 0.406]` and for the standard deviations `[0.229, 0.224, 0.225]`. You'll want to subtract the means from each color channel, then divide by the standard deviation. \n","\n","And finally, PyTorch expects the color channel to be the first dimension but it's the third dimension in the PIL image and Numpy array. You can reorder dimensions using [`ndarray.transpose`](https://docs.scipy.org/doc/numpy-1.13.0/reference/generated/numpy.ndarray.transpose.html). The color channel needs to be first and retain the order of the other two dimensions."]},{"cell_type":"code","execution_count":32,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:51:53.264647Z","iopub.status.busy":"2023-12-30T11:51:53.264021Z","iopub.status.idle":"2023-12-30T11:51:53.272789Z","shell.execute_reply":"2023-12-30T11:51:53.271386Z","shell.execute_reply.started":"2023-12-30T11:51:53.264610Z"},"trusted":true},"outputs":[],"source":["\n","def process_image(image_path):\n","    ''' Scales, crops, and normalizes a PIL image for a PyTorch model,\n","        returns an Numpy array\n","    '''\n","    with Image.open(image_path) as im:\n","        im = im.resize((256,256))\n","        mar_1 = 16 # (256-224)*0.5\n","        mar_2 = 240 #224+16\n","        im = im.crop((mar_1,mar_1,mar_2,mar_2))#l,u,r,l\n","        im = np.array(im)/255\n","        mean = [0.485, 0.456, 0.406]\n","        std = [0.229, 0.224, 0.225]\n","        norm_img = (im - mean)/ std\n","    \n","    return norm_img.transpose((2,0,1))\n","    \n","    "]},{"cell_type":"markdown","metadata":{},"source":["To check your work, the function below converts a PyTorch tensor and displays it in the notebook. If your `process_image` function works, running the output through this function should return the original image (except for the cropped out portions)."]},{"cell_type":"code","execution_count":33,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:51:57.873703Z","iopub.status.busy":"2023-12-30T11:51:57.873297Z","iopub.status.idle":"2023-12-30T11:51:57.881889Z","shell.execute_reply":"2023-12-30T11:51:57.880228Z","shell.execute_reply.started":"2023-12-30T11:51:57.873666Z"},"trusted":true},"outputs":[],"source":["def imshow(image, ax=None, title=None):\n","    \"\"\"Imshow for Tensor.\"\"\"\n","    if ax is None:\n","        fig, ax = plt.subplots()\n","    \n","    # PyTorch tensors assume the color channel is the first dimension\n","    # but matplotlib assumes is the third dimension\n","    image =image.transpose((1, 2, 0))\n","    \n","    # Undo preprocessing\n","    mean = np.array([0.485, 0.456, 0.406])\n","    std = np.array([0.229, 0.224, 0.225])\n","    image = std * image + mean\n","    \n","    # Image needs to be clipped between 0 and 1 or it looks like noise when displayed\n","    image = np.clip(image, 0, 1)\n","    \n","    ax.imshow(image)\n","    \n","    return ax"]},{"cell_type":"markdown","metadata":{},"source":["## Class Prediction\n","\n","Once you can get images in the correct format, it's time to write a function for making predictions with your model. A common practice is to predict the top 5 or so (usually called top-$K$) most probable classes. You'll want to calculate the class probabilities then find the $K$ largest values.\n","\n","To get the top $K$ largest values in a tensor use [`x.topk(k)`](http://pytorch.org/docs/master/torch.html#torch.topk). This method returns both the highest `k` probabilities and the indices of those probabilities corresponding to the classes. You need to convert from these indices to the actual class labels using `class_to_idx` which hopefully you added to the model or from an `ImageFolder` you used to load the data ([see here](#Save-the-checkpoint)). Make sure to invert the dictionary so you get a mapping from index to class as well.\n","\n","Again, this method should take a path to an image and a model checkpoint, then return the probabilities and classes.\n","\n","```python\n","probs, classes = predict(image_path, model)\n","print(probs)\n","print(classes)\n","> [ 0.01558163  0.01541934  0.01452626  0.01443549  0.01407339]\n","> ['70', '3', '45', '62', '55']\n","```"]},{"cell_type":"code","execution_count":34,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:52:00.422407Z","iopub.status.busy":"2023-12-30T11:52:00.421438Z","iopub.status.idle":"2023-12-30T11:52:00.430663Z","shell.execute_reply":"2023-12-30T11:52:00.429605Z","shell.execute_reply.started":"2023-12-30T11:52:00.422362Z"},"trusted":true},"outputs":[],"source":["\n","def predict(image_path, model, topk=5):\n","    ''' Predict the class (or classes) of an image using a trained deep learning model.\n","    '''\n","    \n","    # TODO: Implement the code to predict the class from an image file\n","    image = process_image(image_path)\n","    \n","    image = torch.FloatTensor(image)\n","    image.unsqueeze_(0) # batch dimension\n","    image = image.to(device)\n","    \n","    model.to(device)\n","    model.eval()\n","    with torch.no_grad():\n","        output = model(image)\n","    \n","    prob = torch.exp(output) \n","    top_5_prob, top_5_class = prob.topk(5, dim=1)\n","    \n","    top_5_prob = top_5_prob.squeeze().tolist()\n","    top_5_class = top_5_class.squeeze().tolist()#indexes\n","    \n","    top_5_class_names = [idx_to_class[idx] for idx in top_5_class]\n","    \n","    return top_5_prob, top_5_class_names"]},{"cell_type":"markdown","metadata":{},"source":["## Sanity Checking\n","\n","Now that you can use a trained model for predictions, check to make sure it makes sense. Even if the testing accuracy is high, it's always good to check that there aren't obvious bugs. Use `matplotlib` to plot the probabilities for the top 5 classes as a bar graph, along with the input image. It should look like this:\n","\n","<img src='assets/inference_example.png' width=300px>\n","\n","You can convert from the class integer encoding to actual flower names with the `cat_to_name.json` file (should have been loaded earlier in the notebook). To show a PyTorch tensor as an image, use the `imshow` function defined above."]},{"cell_type":"code","execution_count":37,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:52:03.914424Z","iopub.status.busy":"2023-12-30T11:52:03.913299Z","iopub.status.idle":"2023-12-30T11:52:03.924380Z","shell.execute_reply":"2023-12-30T11:52:03.923135Z","shell.execute_reply.started":"2023-12-30T11:52:03.914383Z"},"trusted":true},"outputs":[],"source":["# TODO: Display an image along with the top 5 classes\n","\n","def check_sanity(test_image_path, model):\n","    probs, classes = predict(test_image_path, model)\n","    print(classes, probs)\n","    \n","    flower_names = [name for name in classes]\n","    \n","    # Creating subplots with matplotlib\n","    plt.figure(figsize=(14,4 ))    \n","    plt.subplot(1,2,1)\n","    plt.axis('off')\n","    plt.title('Image of the Test Flower')\n","    image = process_image(test_image_path)\n","    imshow(image,plt)\n","\n","    plt.subplot(1,2,2)\n","    x_ticks = np.arange(0, 1.2, 0.2)\n","    y_ticks = np.arange(5)\n","    plt.xticks(x_ticks)\n","    plt.yticks(y_ticks,flower_names )\n","    plt.xticks(x_ticks)\n","    plt.title('Sanity check')\n","    plt.xlabel('Probabilities')\n","    plt.ylabel('Flower Types')\n","    plt.barh(y_ticks, probs, align='center')    \n","\n","    plt.show()\n","    truth = test_image_path.split('/')[-2]\n","    print('ground truth for the image is :',idx_to_class[model.class_to_idx[truth]])\n"]},{"cell_type":"code","execution_count":38,"metadata":{"execution":{"iopub.execute_input":"2023-12-30T11:52:48.037889Z","iopub.status.busy":"2023-12-30T11:52:48.036731Z","iopub.status.idle":"2023-12-30T11:52:48.086714Z","shell.execute_reply":"2023-12-30T11:52:48.085351Z","shell.execute_reply.started":"2023-12-30T11:52:48.037821Z"},"trusted":true},"outputs":[{"name":"stdout","output_type":"stream","text":["['camellia', 'mallow', 'lenten rose', 'rose', 'carnation'] [0.9946340918540955, 0.0036088041961193085, 0.0013992780586704612, 0.0001696998078841716, 0.00010157281212741509]\n"]},{"data":{"image/png":"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","text/plain":["<Figure size 1400x400 with 2 Axes>"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["ground truth for the image is : camellia\n"]}],"source":["image_path = '/Users/dalex/Udacity/aipnd-project/flower_data/test/96/image_07597.jpg'\n","\n","check_sanity(image_path, loaded_model)"]},{"cell_type":"markdown","metadata":{},"source":["## Reminder for Workspace users\n","If your network becomes very large when saved as a checkpoint, there might be issues with saving backups in your workspace. You should reduce the size of your hidden layers and train again. \n","    \n","We strongly encourage you to delete these large interim files and directories before navigating to another page or closing the browser tab."]},{"cell_type":"code","execution_count":null,"metadata":{},"outputs":[],"source":["# TODO remove .pth files or move it to a temporary `~/opt` directory in this Workspace\n","!rm '/Users/dalex/Udacity/aipnd-project/checkpoint_5_ep_densenet_121_experimentation.pth'\n","!rm '/Users/dalex/Udacity/aipnd-project/checkpoint_5_ep_resnet50_experimentation.pth'"]},{"cell_type":"code","execution_count":null,"metadata":{},"outputs":[],"source":["#code for retraining a loaded model\n","epochs = 10\n","new_trained_model, train_losses, valid_losses, val_acc = train_model(\n","    loaded_model, epochs,trainloaders, validloaders,criterion, optimizer, scheduler,gpu ='gpu') "]},{"cell_type":"code","execution_count":null,"metadata":{},"outputs":[],"source":["#references followed are:\n","#https://towardsdatascience.com/flower-species-classifier-c8893030cd90\n","#https://github.com/dkhundley/flower-image-classifier/tree/master\n","#https://github.com/ardaerendogru/Flower-Breed-Classifier\n"]}],"metadata":{"kaggle":{"accelerator":"nvidiaTeslaT4","dataSources":[{"datasetId":2921509,"sourceId":5033713,"sourceType":"datasetVersion"},{"datasetId":3041472,"sourceId":5227346,"sourceType":"datasetVersion"},{"datasetId":4240296,"sourceId":7307965,"sourceType":"datasetVersion"}],"dockerImageVersionId":30408,"isGpuEnabled":true,"isInternetEnabled":true,"language":"python","sourceType":"notebook"},"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.10.4"}},"nbformat":4,"nbformat_minor":4}