Changing the hashing methodology for cache folder creation of models. #481
Conversation
… QNN compilation not included yet. Cache folder mechanism has been modified to have a parent directory for a model based on the architecture that we retrieve from the model config. The hash calculation for the ONNX export now incorporates all model kwargs as well as export kwargs and parameters. the parameters that were used to create the hash also gets dumped as a serialized JSON file in the ONNX folder, the same happens for the compile parameters inside the respective qpc folder. Signed-off-by: Dhiraj Kumar Sah <[email protected]>
quic-dhirajku
requested review from
quic-rishinr,
ochougul,
quic-hemagnih and
quic-amitraj
as code owners
| @@ -5,7 +5,7 @@ | |||
| # | |||
| # ---------------------------------------------------------------------------- | |||
|
|
|||
| import hashlib | |||
| # import hashlib | |||
There was a problem hiding this comment.
commented code.
Make sure that commented code is not there in ready to review PRs.
| self.model_params.update(kwargs) | ||
| self.model_params["config"] = self.model.config.to_diff_dict() | ||
| self.model_params["_transform_names"] = self._transform_names() | ||
| self.compile_params = {} |
There was a problem hiding this comment.
initialize this only when compile is called.
No point in creating this dictionary if user not calling compile.
| self.model_params = {} | ||
| self.model_params.update(kwargs) |
There was a problem hiding this comment.
Better to do self.model_params = copy.deepcopy(kwargs)
This lets other methods mutate kwargs.
Otherwise we would need to ensure that no other method mutates the kwargs.
| if export_kwargs is not None: | ||
| self.model_params.update(export_kwargs) | ||
| if onnx_transform_kwargs is not None: | ||
| self.model_params.update(onnx_transform_kwargs) |
There was a problem hiding this comment.
One liners are better
self.model_params.update(export_kwargs) if export_kwargs is not None else None
self.model_params.update(onnx_transform_kwargs) if export_kwargs is not None else None
| self.model_params["output_names"] = output_names | ||
| self.model_params["dynamic_axes"] = dynamic_axes |
There was a problem hiding this comment.
Better to keep them in one more level as
self.model_params["export_params"] = export_params
And add all exports params in export_params which is another dict.
Makes the dumped JSON readable by user.
| model_params_json = export_dir / "model_params.json" | ||
| with open(model_params_json, "w") as fp: | ||
| json.dump( | ||
| { | ||
| "model_params": [ | ||
| {k: make_serializable(self.model_params[k]) for k in sorted(self.model_params.keys())} | ||
| ] | ||
| }, | ||
| fp, | ||
| indent=4, | ||
| ) |
There was a problem hiding this comment.
Dumping should happen after export.
If model errors out during export and we still dump the json, it does not make sense
|
|
||
| self.pretrained_model_name_or_path = kwargs.get("pretrained_model_name_or_path", None) | ||
| # self.pretrained_model_name_or_path = kwargs.get("pretrained_model_name_or_path", None) |
Detaching hash function for model cache path calculation. changes for QNN compilation not included yet.
Cache folder mechanism has been modified to have a parent directory for a model based on the architecture that we retrieve from the model config. The hash calculation for the ONNX export now incorporates all model kwargs as well as export kwargs and parameters. the parameters that were used to create the hash also gets dumped as a serialized JSON file in the ONNX folder, the same happens for the compile parameters inside the respective qpc folder.