RuntimeError: Error(s) in loading state_dict for QuantizedFasterRCNN: #1123
Labels
Comments
|
how did you get the quantized model? |
yanbing-j
pushed a commit
to yanbing-j/ao
that referenced
this issue
Dec 9, 2024
This PR makes torchchat support multi-modality model definition and constructions. To show our power in multi-modality area, we integrate flamingo component into our system. Note that this is only for bare-minimum support for model definition. Please check openai_api_multimodal branch for e2e, and pytorch#1123 (comment) for better structure and llama3.1 support
yanbing-j
pushed a commit
to yanbing-j/ao
that referenced
this issue
Dec 9, 2024
* added model source and type for torchtune flamingo support * added model source and type for torchtune flamingo support * grab missing enum * fix ModelArgs init * create init func for ModelArgs for BC * update pipeline for ModleSource and ModelType * revert lintrunner update on ET * introduce flamingo modules form torchtune * back up to move to linux * mitigate building issue * pass local test * structual model builder * update torchtune address * update install requirement * support new torchtune flamingo component * specific version for vision and ao * unify text-only model generation pipeline * convert installation back and bypass torchtune * restructual model definition * update exportation variable name * remove redunctant function * 1/n torchtune 3.1 8b * installation update * torchtune 3.1 8b / 30b * bring torchchat llama3.1 back * bring tok vali back to torchchat model + revert install_requirements.sh * solve bugs related to tt model support * bypass torchtune import issue * solve Jack's wonderful comments * remveo extra dot * add type.Callable * fix torchchat typos * solve bug when args.model is None * support builder_args.params_table is None * remove all .DS_Store * bring gguf back * remove reduntant updates * bring checkpoint back * debug * debug * debug * new factory func to produce Model from modelargs * solve comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
I have fine-tuned and quantized faster_rcnn model for the use but when I load it...
RuntimeError Traceback (most recent call last)
in <cell line: 10>()
8
9 # Load the quantized model weights
---> 10 quantized_model.load_state_dict(torch.load('/content/drive/MyDrive/Aditya Pandey/global_wheat_detection-2-20241004T044942Z-001/global_wheat_detection-2/Copy of faster_rcnn_40MB.pth', map_location=torch.device('cpu')))
11
12 # Set the model to evaluation mode
/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py in load_state_dict(self, state_dict, strict, assign)
2213
2214 if len(error_msgs) > 0:
-> 2215 raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
2216 self.class.name, "\n\t".join(error_msgs)))
2217 return _IncompatibleKeys(missing_keys, unexpected_keys)
RuntimeError: Error(s) in loading state_dict for QuantizedFasterRCNN:
Missing key(s) in state_dict: "model.backbone.body.conv1.weight", "model.backbone.body.bn1.weight", "model.backbone.body.bn1.bias", "model.backbone.body.bn1.running_mean", "model.backbone.body.bn1.running_var", "model.backbone.body.layer1.0.conv1.weight", "model.backbone.body.layer1.0.bn1.weight", "model.backbone.body.layer1.0.bn1.bias", "model.backbone.body.layer1.0.bn1.running_mean", "model.backbone.body.layer1.0.bn1.running_var", "model.backbone.body.layer1.0.conv2.weight", "model.backbone.body.layer1.0.bn2.weight", "model.backbone.body.layer1.0.bn2.bias", "model.backbone.body.layer1.0.bn2.running_mean", "model.backbone.body.layer1.0.bn2.running_var", "model.backbone.body.layer1.0.conv3.weight", "model.backbone.body.layer1.0.bn3.weight", "model.backbone.body.layer1.0.bn3.bias", "model.backbone.body.layer1.0.bn3.running_mean", "model.backbone.body.layer1.0.bn3.running_var", "model.backbone.body.layer1.0.downsample.0.weight", "model.backbone.body.layer1.0.downsample.1.weight", "model.backbone.body.layer1.0.downsample.1.bias", "model.backbone.body.layer1.0.downsample.1.running_mean", "model.backbone.body.layer1.0.downsample.1.running_var", "model.backbone.body.layer1.1.conv1.weight", "model.backbone.body.layer1.1.bn1.weight", "model.backbone.body.layer1.1.bn1.bias", "model.backbone.body.layer1.1.bn1.running_mean", "model.backbone.body.layer1.1.bn1.running_var", "model.backbone.body.layer1.1.conv2.weight", "model.backbone.body.layer1.1.bn2.weight", "model.backbone.body.lay...
Unexpected key(s) in state_dict: "faster_rcnn.backbone.body.conv1.weight", "faster_rcnn.backbone.body.conv1.bias", "faster_rcnn.backbone.body.conv1.scale", "faster_rcnn.backbone.body.conv1.zero_point", "faster_rcnn.backbone.body.bn1.weight", "faster_rcnn.backbone.body.bn1.bias", "faster_rcnn.backbone.body.bn1.running_mean", "faster_rcnn.backbone.body.bn1.running_var", "faster_rcnn.backbone.body.layer1.0.conv1.weight", "faster_rcnn.backbone.body.layer1.0.conv1.bias", "faster_rcnn.backbone.body.layer1.0.conv1.scale", "faster_rcnn.backbone.body.layer1.0.conv1.zero_point", "faster_rcnn.backbone.body.layer1.0.bn1.weight", "faster_rcnn.backbone.body.layer1.0.bn1.bias", "faster_rcnn.backbone.body.layer1.0.bn1.running_mean", "faster_rcnn.backbone.body.layer1.0.bn1.running_var", "faster_rcnn.backbone.body.layer1.0.conv2.weight", "faster_rcnn.backbone.body.layer1.0.conv2.bias", "faster_rcnn.backbone.body.layer1.0.conv2.scale", "faster_rcnn.backbone.body.layer1.0.conv2.zero_point", "faster_rcnn.backbone.body.layer1.0.bn2.weight", "faster_rcnn.backbone.body.layer1.0.bn2.bias", "faster_rcnn.backbone.body.layer1.0.bn2.running_mean", "faster_rcnn.backbone.body.layer1.0.bn2.running_var", "faster_rcnn.backbone.body.layer1.0.conv3.weight", "faster_rcnn.backbone.body.layer1.0.conv3.bias", "faster_rcnn.backbone.body.layer1.0.conv3.scale", "faster_rcnn.backbone.body.layer1.0.conv3.zero_point", "faster_rcnn.backbone.body.layer1.0.bn3.weight", "faster_rcnn.backbone.body.layer1.0.bn3.bias", "fas...
The text was updated successfully, but these errors were encountered: