simple-shapes-dataset-inference.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "75a2017e-7de7-4568-87dd-23bf4f93a0cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "from typing import cast\n",
    "\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import torch\n",
    "from lightning.pytorch import seed_everything\n",
    "from shimmer.modules.global_workspace import (\n",
    "    GlobalWorkspaceFusion,\n",
    ")\n",
    "from shimmer_ssd import DEBUG_MODE, PROJECT_DIR\n",
    "from shimmer_ssd.config import DomainModelVariantType, LoadedDomainConfig, load_config\n",
    "from shimmer_ssd.dataset.pre_process import TokenizeCaptions\n",
    "from shimmer_ssd.logging import batch_to_device\n",
    "from shimmer_ssd.modules.domains import load_pretrained_domains\n",
    "from shimmer_ssd.modules.domains.visual import VisualLatentDomainModule\n",
    "from simple_shapes_dataset import (\n",
    "    SimpleShapesDataModule,\n",
    "    get_default_domains,\n",
    ")\n",
    "from tokenizers.implementations.byte_level_bpe import ByteLevelBPETokenizer\n",
    "from torchvision.utils import make_grid\n",
    "\n",
    "matplotlib.use(\"ipympl\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "686100fa-404a-4fba-80f7-bfc17d4d40d6",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Seed set to 0\n",
      "/home/bdevillers/.cache/pypoetry/virtualenvs/shimmer-ssd-s1RBJB6U-py3.11/lib/python3.11/site-packages/lightning/pytorch/core/saving.py:191: Found keys that are in the model state dict but not in the checkpoint: ['logit_scale']\n"
     ]
    }
   ],
   "source": [
    "seed = 0\n",
    "dataset_path = \"\"\n",
    "\n",
    "seed_everything(seed, workers=True)\n",
    "\n",
    "# Let's define the proportion of each groups of domain.\n",
    "# Here, we say that we will use 100% of unimodal text domain samples\n",
    "# 100% of the unimodal visual samples\n",
    "# and 100% of the paired text-visual samples.\n",
    "domain_proportion = {\n",
    "    frozenset([\"t\"]): 1.0,\n",
    "    frozenset([\"v\"]): 1.0,\n",
    "    frozenset([\"t\", \"v\"]): 1.0,\n",
    "}\n",
    "\n",
    "# Then we get the data domain that will load the data for our two different domain\n",
    "# formats:\n",
    "# - `t` will load the text (with bert representations of the sentences\n",
    "# - `v_latents` will load pre-saved latent representations of the visual VAE.\n",
    "# You could also use `v` to load the images directly. Using `v_latents` will provide\n",
    "# faster training as we can skip the VAE encoding.\n",
    "domain_classes = get_default_domains([\"v_latents\", \"t\"])\n",
    "\n",
    "# Let's define the locations of the `vocab.json` and `merges.txt` files for the\n",
    "# text tokenizer.\n",
    "vocab_path = \"/home/bdevillers/projects/shimmer-ssd/tokenizer/vocab.json\"\n",
    "merges_path = \"/home/bdevillers/projects/shimmer-ssd/tokenizer/merges.txt\"\n",
    "seq_length = 64  # we will use a sequence length of 64\n",
    "\n",
    "# And define a transform for the text domain to tokenize the text string\n",
    "additional_transforms = {\"t\": [TokenizeCaptions(vocab_path, merges_path, seq_length)]}\n",
    "tokenizer = ByteLevelBPETokenizer(vocab_path, merges_path)\n",
    "\n",
    "# here are some additional arguments used by the data domain classes\n",
    "domain_data_args = {\n",
    "    # where to load the pre-saved latent reprensentations\n",
    "    \"v_latents\": {\"presaved_path\": \"vae_v_shimmer.npy\"},\n",
    "    # bert model to use\n",
    "    \"t\": {\"latent_filename\": \"bert-base-uncased\"},\n",
    "}\n",
    "\n",
    "batch_size = 1024\n",
    "\n",
    "# initilize the lightning data module\n",
    "data_module = SimpleShapesDataModule(\n",
    "    config.dataset.path,\n",
    "    domain_classes,\n",
    "    domain_proportion,\n",
    "    batch_size,\n",
    "    seed=seed,\n",
    "    domain_args=domain_data_args,\n",
    "    additional_transforms=additional_transforms,\n",
    "    max_train_size=500_000,\n",
    ")\n",
    "\n",
    "# We need to load pretrained unimodal domains.\n",
    "# Here we define where are stored the domain checkpoints.\n",
    "domains = [\n",
    "    LoadedDomainConfig(domain_type=DomainModelVariantType.v_latents,\n",
    "        checkpoint_path=Path(\n",
    "            \"/home/bdevillers/projects/shimmer-ssd/checkpoints/pretrained/domain_v.ckpt\"\n",
    "        ),\n",
    "    ),\n",
    "    LoadedDomainConfig(\n",
    "        domain_type=DomainModelVariantType.t,\n",
    "        checkpoint_path=Path(\n",
    "            \"/home/bdevillers/projects/shimmer-ssd/checkpoints/pretrained/domain_t.ckpt\"\n",
    "        ),\n",
    "    ),\n",
    "]\n",
    "\n",
    "global_workspace_dim = 16\n",
    "encoders_hidden_dim = 32\n",
    "encoders_n_layers = 3\n",
    "decoders_hidden_dim = 32\n",
    "decoders_n_layers = 3\n",
    "\n",
    "# we load the pretrained domain modules and define the associated GW encoders and\n",
    "# decoders\n",
    "domain_modules, gw_encoders, gw_decoders = load_pretrained_domains(\n",
    "    domains,\n",
    "    global_workspace_dim,\n",
    "    encoders_hidden_dim,\n",
    "    encoders_n_layers,\n",
    "    decoders_hidden_dim,\n",
    "    decoders_n_layers,\n",
    ")\n",
    "\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "\n",
    "# And now we load the GW checkpoint\n",
    "checkpoint = \"/home/bdevillers/projects/shimmer-ssd/checkpoints/shimmer-simple-shapes-caer5sh9/epoch=1232.ckpt\"\n",
    "global_workspace = GlobalWorkspaceFusion.load_from_checkpoint(\n",
    "    checkpoint,\n",
    "    domain_mods=domain_modules,\n",
    "    gw_encoders=gw_encoders,\n",
    "    gw_decoders=gw_decoders,\n",
    ")\n",
    "global_workspace.to(device)\n",
    "\n",
    "# Extract the `v_latents` domain module\n",
    "visual_module = cast(\n",
    "    VisualLatentDomainModule, global_workspace.domain_mods[\"v_latents\"]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "0ec5bcca-7c39-4d18-b385-d006cebd20f8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys([frozenset({'v_latents'}), frozenset({'t'}), frozenset({'t', 'v_latents'})])\n"
     ]
    }
   ],
   "source": [
    "# Let's get 32 samples from our dataset\n",
    "train_samples = data_module.get_samples(\"train\", 32)\n",
    "train_samples = batch_to_device(train_samples, device)\n",
    "print(train_samples.keys())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3f2f7e6c-15ad-4211-b491-f3257919faf6",
   "metadata": {},
   "source": [
    "This contains three groups: 2 unimodal data groups, and one paired group.\n",
    "This corresponds to the groups defined in `domain_proportions`.\n",
    "<div class=\"alert alert-info\">\n",
    "NOTE:\n",
    "This is only the case for the train split. The train and test splits only have the paired data group as everything is paired.   \n",
    "</div>\n",
    "\n",
    "In the following, we will focus on the paired data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "880e52bc-627e-4292-958b-632715aa076b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys(['v_latents', 't'])\n"
     ]
    }
   ],
   "source": [
    "train_paired_samples = train_samples[frozenset([\"v_latents\", \"t\"])]\n",
    "print(train_paired_samples.keys())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dd450237-cea5-4ff1-831a-e785126a5e7a",
   "metadata": {},
   "source": [
    "This group contains two items: one of each domain. Each item in the domain is paired\n",
    "Let's look at the images first."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "c3e9c3fe-d665-4de3-8d04-48d5b0c536ea",
   "metadata": {},
   "outputs": [],
   "source": [
    "# `train_paired_samples` contains the latent representation, we can use `decode_images`\n",
    "# of the visual module to get back the actual images:\n",
    "train_images = visual_module.decode_images(train_paired_samples[\"v_latents\"])\n",
    "\n",
    "\n",
    "def plot_image_grid(samples: torch.Tensor, nrow=4):\n",
    "    grid = (\n",
    "        make_grid(samples, nrow=nrow, pad_value=1)\n",
    "        .permute(1, 2, 0)\n",
    "        .detach()\n",
    "        .cpu()\n",
    "        .numpy()\n",
    "    )\n",
    "    plt.imshow(grid)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "5d2a5fb5-7ec8-4bb1-a55e-3ac41596ee52",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([32, 13])"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_paired_samples[\"v_latents\"].size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "52e0e0e2-72c9-4ae9-ac42-0abf99978b83",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ce0b60423dfe4f26b4bd539ab9814317",
       "version_major": 2,
       "version_minor": 0
      },
      "image/png": "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",
      "text/html": [
       "\n",
       "            <div style=\"display: inline-block;\">\n",
       "                <div class=\"jupyter-widgets widget-label\" style=\"text-align: center;\">\n",
       "                    Figure\n",
       "                </div>\n",
       "                <img src='data:image/png;base64,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' width=640.0/>\n",
       "            </div>\n",
       "        "
      ],
      "text/plain": [
       "Canvas(toolbar=Toolbar(toolitems=[('Home', 'Reset original view', 'home', 'home'), ('Back', 'Back to previous …"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_image_grid(train_images)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f092ed7b-1f31-416c-b2af-2dd41aa92877",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'bert': tensor([[-0.0295,  0.4828, -0.6249,  ..., -0.6715, -0.0847,  0.2428],\n",
       "         [-0.3529, -0.1964,  0.5454,  ..., -1.8058,  0.5423, -0.9121],\n",
       "         [ 0.8766, -0.1070,  1.7887,  ...,  1.1233,  0.4867, -0.1705],\n",
       "         ...,\n",
       "         [-1.9229,  2.2754, -2.2220,  ...,  1.3829, -2.1496,  0.0222],\n",
       "         [-0.3343, -0.1584,  1.2444,  ..., -0.7224,  0.3656,  1.3582],\n",
       "         [ 1.2014,  0.4460,  0.0669,  ...,  0.9270,  0.4732, -0.1380]],\n",
       "        device='cuda:0'),\n",
       " 'tokens': tensor([[314, 346, 436,  ...,   0,   0,   0],\n",
       "         [428, 265, 261,  ...,   0,   0,   0],\n",
       "         [314, 346, 265,  ...,   0,   0,   0],\n",
       "         ...,\n",
       "         [429, 265, 261,  ...,   0,   0,   0],\n",
       "         [295, 265, 261,  ...,   0,   0,   0],\n",
       "         [ 35, 358, 383,  ...,   0,   0,   0]], device='cuda:0')}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_paired_samples[\"t\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "5beac4be-ee02-4849-937c-77646922b714",
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_text_samples(samples: dict[str, torch.Tensor]) -> list[str]:\n",
    "    return tokenizer.decode_batch(\n",
    "        samples[\"tokens\"].detach().cpu().tolist(),\n",
    "        skip_special_tokens=True,\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "c54275e7-33ab-42a2-8651-040d766db19d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 The image contains a thistle colored tiny arrow-shaped polygon. It is in the middle, slightly left. It is pointing towards the southwest.\n",
      "2 This is a big dark slate blue colored egg, it is located at the upper side right side, it is pointing to the right.\n",
      "3 The image is a tiny forest green colored object located at the bottom left side, and is pointing towards the northeast, and it looks like a trapezoidal shape.\n",
      "4 The image contains a medium sized object in silver pointing towards the south-southeast, and is located slightly lower side, at the center, it looks like a guitar pick.\n",
      "5 It is a tiny dark slate grey object at the bottom, on the left side and is pointing northwest. It looks like an oval shaped structure.\n",
      "6 A small dark goldenrod isosceles triangle, and is at the middle, and is pointing to the west-southwest.\n",
      "7 The image contains a medium dark grey trapezoidal shape and it is pointing towards the top-right corner and it is at the bottom, at the left.\n",
      "8 This is an average sized light pink colored shape pointing to the bottom and it is in the middle. It looks like an isosceles triangle.\n",
      "9 A small dark sea green colored four-sided shape, and it is pointing towards the right top-right, and it is at the top right.\n",
      "10 A kind of dark blue big deformed square shape, pointing to the bottom-left corner, located at the bottom center.\n",
      "11 The image is a trapezoidal shape, and it is medium sized. It is in indigo, and it is located in the middle right, at the very right side, and it is pointing east.\n",
      "12 The image is an average sized shape pointing towards the left side. It is at the lower side, slightly right. It is medium slate blue colored, it looks like a triangle.\n",
      "13 A kind of average sized shape in dark olive green color located slightly top, at the center and it is pointing northwest, it looks like an egg.\n",
      "14 There is a tiny violet object pointing to the bottom, located at the top left side, it looks like a triangle.\n",
      "15 There is an isosceles triangle and is small, it is saddle brown and is pointing south-southwest and is located at the upper side left side.\n",
      "16 The image is a tiny shape in silver color pointing to the top top-right, at the very lower side, on the right side. It looks like a deformed square shape.\n",
      "17 The image contains a large dark slate blue colored shape pointing to the top-right and it is located at the very bottom, at the center and looks like an oval shaped structure.\n",
      "18 A kind of plum medium sized guitar pick and is located at the lower side right side and is pointing towards the northwest.\n",
      "19 A kind of grey isosceles triangle that has round corners, pointing north-northwest, located in the middle left, small.\n",
      "20 A kind of water droplet and is medium sized. It is in olive drab and is located top and is pointing to the top-left corner.\n",
      "21 The image contains a medium sized grey shape at the very upper side left, on the very left, it is pointing towards the bottom-right corner. It looks like a kite.\n",
      "22 There is a bullet, it is tiny, and is in indian red, it is at the middle left, it is pointing to the northeast.\n",
      "23 The image represents a large shape in pale green pointing towards the east. It is located at the very upper side, slightly left. It looks like a kite.\n",
      "24 The image is a big maroon deformed square shape, and is located at the center, at the left side, and is pointing towards the south.\n",
      "25 The image is a khaki oval shaped structure. It is medium. It is located at the top, at the right. It is pointing to the bottom-left corner.\n",
      "26 A medium sized light steel blue triangle, it is pointing to the left side, it is located at the top center.\n",
      "27 The image represents a little object at the bottom middle, pointing towards the west-southwest, in dark blue, it looks like an arrow-shaped polygon.\n",
      "28 The image represents a beige triangle, and is at the very lower side, slightly right, and is pointing to the top, and is tiny.\n",
      "29 A quadrilateral, big and is in silver color, at the bottom, at the left side, pointing to the bottom-left.\n",
      "30 There is a guitar pick, and is little. It is linen, and is at the bottom right, at the very right side, and is pointing towards the top.\n",
      "31 It is a medium orchid colored water droplet and it is pointing south-southeast and it is located at the upper side left and it is small.\n",
      "32 A tiny object located at the center left side. It is pointing northeast. It is dark slate grey. It looks like a triangle.\n"
     ]
    }
   ],
   "source": [
    "train_text_samples = get_text_samples(train_paired_samples[\"t\"])\n",
    "for k, text in enumerate(train_text_samples, 1):\n",
    "    print(k, text.replace(\"<pad>\", \"\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "0baf3400-f305-4d28-919a-a884abd597fd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([32, 64])\n"
     ]
    }
   ],
   "source": [
    "latent_unimodal_text = global_workspace.encode_domain(train_paired_samples[\"t\"], \"t\")\n",
    "print(latent_unimodal_text.size())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "52acf10d-3c2f-47d6-97d0-0411812b9659",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "torch.Size([32, 16])\n",
      "torch.Size([32, 16])\n"
     ]
    }
   ],
   "source": [
    "t_gw_latent = global_workspace.gw_mod.encode({\"t\": latent_unimodal_text})\n",
    "print(t_gw_latent[\"t\"].size())\n",
    "t_gw_latent_fused = global_workspace.gw_mod.fuse(\n",
    "    t_gw_latent, {\"t\": torch.ones(32).to(device)}\n",
    ")\n",
    "print(t_gw_latent_fused.size())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "4dbce724-f6c6-4b85-9bdd-4b035b40c2bc",
   "metadata": {},
   "outputs": [],
   "source": [
    "decoded_latents = global_workspace.gw_mod.decode(\n",
    "    t_gw_latent_fused, domains=[\"v_latents\"]\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "e85b64d4-a0b6-4b02-9ed8-5ec7e19ab66d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "torch.Size([32, 12])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "decoded_latents[\"v_latents\"].size()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "a44a7a8c-3335-4440-bd75-2613d0bdc81f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "975bfc9c7f3b4350b88773510cb54dcc",
       "version_major": 2,
       "version_minor": 0
      },
      "image/png": "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",
      "text/html": [
       "\n",
       "            <div style=\"display: inline-block;\">\n",
       "                <div class=\"jupyter-widgets widget-label\" style=\"text-align: center;\">\n",
       "                    Figure\n",
       "                </div>\n",
       "                <img src='data:image/png;base64,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' width=640.0/>\n",
       "            </div>\n",
       "        "
      ],
      "text/plain": [
       "Canvas(toolbar=Toolbar(toolitems=[('Home', 'Reset original view', 'home', 'home'), ('Back', 'Back to previous …"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "decoded_images = visual_module.decode_images(\n",
    "    decoded_latents[\"v_latents\"], remove_last_dim=False\n",
    ")\n",
    "plot_image_grid(decoded_images)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "85bea5b7-a747-40f6-a9e2-897c2521e9a4",
   "metadata": {},
   "outputs": [],
   "source": [
    "latent_unimodal_images = global_workspace.encode_domains(train_samples)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "b245799a-e028-4d5b-af3f-54ecd679f871",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "dict_keys([frozenset({'t'})])\n"
     ]
    }
   ],
   "source": [
    "t_ = frozenset([\"t\"])\n",
    "unimodal_train_samples = {t_: train_samples[t_]}\n",
    "latent_unimodal_images = global_workspace.encode_domains(unimodal_train_samples)\n",
    "print(latent_unimodal_images.keys())\n",
    "gw_latents = global_workspace.encode(latent_unimodal_images)\n",
    "gw_latents_fused = global_workspace.fuse(\n",
    "    gw_latents, {t_: {\"t\": torch.ones(32).to(device)}}\n",
    ")\n",
    "decoded_latents = global_workspace.decode(gw_latents_fused)\n",
    "reconstructed_domains = global_workspace.decode_domains(decoded_latents)\n",
    "reconstructed_texts = reconstructed_domains[t_][\"t\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "4e73a2f3-536c-400f-996f-f42e69063086",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dict_keys(['token_dist', 'tokens'])"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reconstructed_texts.keys()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "c311bc41-ce50-4e12-9c16-aa790f903b6b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1 The image contains a thistle tiny arrow-shaped polygon, and it is pointing towards the bottom-left, and it is located in the middle, slightly left.\n",
      "2 The image is a large slate blue colored shape pointing to the right side, and is located at the upper side right side, and looks like a bullet.\n",
      "3 This is a tiny forest green colored object pointing towards the northeast, and is located at the bottom left side, and looks like a four-sided shape.\n",
      "4 The image contains a medium sized silver colored object pointing towards the south-southeast, and it is located slightly lower side, in the middle, it looks like a bullet.\n",
      "5 There is a tiny dark slate grey object pointing to the northwest, it is at the bottom, on the left side, and looks like a bullet.\n",
      "6 A kind of small shape in dark goldenrod color pointing to the west-southwest, and is located at the center, and looks like an isosceles triangle.\n",
      "7 The image contains a medium sized dark grey colored shape pointing towards the top-right corner, it is located at the lower side, on the left side, and looks like a kite.\n",
      "8 The image is a light pink colored medium sized isosceles triangle, it is pointing to the south, it is located in the center.\n",
      "9 A kind of little dark sea green colored shape pointing towards the right top-right corner, and it is located at the top right, and looks like a kite.\n",
      "10 A kind of large shape in dark slate blue color pointing southwest, and is at the bottom center, and looks like a quadrilateral.\n",
      "11 This is a medium sized shape in indigo pointing east, and it is located at the middle right, at the very right, and looks like a quadrilateral.\n",
      "12 This is a medium sized shape in medium slate blue color pointing towards the left side, and is at the bottom, slightly right, and looks like an isosceles triangle.\n",
      "13 A kind of medium sized shape in dark olive green color pointing to the northwest, and it is slightly top, in the middle, and it looks like a bullet.\n",
      "14 It is a tiny object in medium orchid color pointing to the bottom, and is at the upper side left, and looks like an isosceles triangle.\n",
      "15 There is a triangle, and is small, and it is saddle brown colored, and is pointing south-southwest, and is at the upper side left side.\n",
      "16 The image is a tiny shape in silver color pointing to the top top-right corner, and it is at the very lower side, on the right side, and looks like a kite.\n",
      "17 The image contains a large dark slate blue colored shape pointing to the top-right corner, and it is at the very lower side, at the center, and it looks like a guitar pick.\n",
      "18 A kind of medium sized plum colored shape pointing towards the northwest, and it is located at the lower side right side, and it looks like a bullet.\n",
      "19 A kind of grey colored bullet, it is pointing north-northwest, it is located in the middle left, it is little.\n",
      "20 A kind of medium sized shape in dark green color pointing to the top-left corner, and is located at the top center, and looks like a guitar pick.\n",
      "21 The image contains a medium sized grey colored shape pointing towards the bottom-left corner, it is at the very upper side left, on the very left side, and looks like a quadrilateral.\n",
      "22 It is a dark salmon colored tiny bullet, it is pointing northeast, it is located in the middle left.\n",
      "23 The image represents a large shape in lime green color pointing towards the east, and it is located at the very upper side, slightly left, and looks like a kite.\n",
      "24 This is a large maroon shape pointing towards the south, it is located at the middle, at the left side, and looks like a quadrilateral.\n",
      "25 This is a khaki colored medium sized guitar pick, and it is pointing to the bottom-left corner, and it is located at the top, at the right side.\n",
      "26 A kind of medium sized object in light steel blue color pointing to the left side, it is located at the top middle, and looks like an isosceles triangle.\n",
      "27 The image represents a little object in blue pointing towards the west-southwest, and it is at the lower side center, and it looks like a kite.\n",
      "28 The image represents a beige colored triangle, and it is tiny, and it is pointing to the top, and it is at the very lower side, slightly right.\n",
      "29 A kind of large shape in silver color pointing southeast, and it is at the lower side, at the left side, and looks like a quadrilateral.\n",
      "30 There is an egg, it is little, and it is in ivory color, it is pointing towards the bottom bottom-right corner, it is at the very top, slightly right.\n",
      "31 It is a medium orchid colored egg, it is pointing south-southeast, it is at the upper side left, it is small.\n",
      "32 A tiny object in dark slate grey pointing northeast, and is located in the middle left side, and looks like a triangle.\n"
     ]
    }
   ],
   "source": [
    "reconstructed_text_strings = get_text_samples(reconstructed_texts)\n",
    "for k, text in enumerate(reconstructed_text_strings, 1):\n",
    "    print(k, text.replace(\"<pad>\", \"\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e0dce11f-ced5-44ac-860e-8da85ed53cb0",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}