pt mobile script and optimize recipe #1193
Conversation
recipes_source/fuse.rst
Outdated
| prepare_save(model_fused, True) | ||
|
|
||
|
|
||
| The outputs of the original model and its fused version will be: |
There was a problem hiding this comment.
I had to read this line twice - when I see "outputs of the model", I think of predictions, but this is referring to the outputs of the print() statements above. For clarity, I'd move the two print statements to the end of the code section above, and change "outputs" to "structures" or something similar.
|
|
||
| import torchvision | ||
|
|
||
| model_quantized = torchvision.models.quantization.mobilenet_v2(pretrained=True, quantize=True) |
There was a problem hiding this comment.
Does it make sense to demonstrate post-training quantization as a separate step? This would be valuable in the case that the user's existing model is not already quantized.
There was a problem hiding this comment.
Yes it does but the Quantization recipe covers the info in details.
| import torch | ||
|
|
||
| dummy_input = torch.rand(1, 3, 224, 224) | ||
| torchscript_model = torch.jit.trace(model_quantized, dummy_input) |
There was a problem hiding this comment.
It would be good to warn the user that torch.jit.trace() only scripts the code path executed during the trace, so it will not work properly for models that include decision branches.
|
|
||
| import torchvision | ||
|
|
||
| model_quantized = torchvision.models.quantization.mobilenet_v2(pretrained=True, quantize=True) |
There was a problem hiding this comment.
(I mentioned this separately in the Android recipe) Does it make sense to demonstrate post-training quantization as a separate step? This would be valuable in the case that the user's existing model is not already quantized.
| import torch | ||
|
|
||
| dummy_input = torch.rand(1, 3, 224, 224) | ||
| torchscript_model = torch.jit.trace(model_quantized, dummy_input) |
There was a problem hiding this comment.
(I mentioned this separately in the Android recipe) It would be good to warn the user that torch.jit.trace() only scripts the code path executed during the trace, so it will not work properly for models that include decision branches.
|
|
||
| * Run `pod install` from the Terminal and then open your project's xcworkspace file; | ||
|
|
||
| * Drag and drop the two files `TorchModule.h` and `TorchModule.mm` to your project. If your project is Swift based, a message box with the title "Would you like to configure an Objective-C bridging header?" will show up; click the "Create Bridging Header" button to create a Swift to Objective-c bridging header file, and add `#import "TorchModule.h"` to the header file `<your_project_name>-Bridging-Header.h`; |
There was a problem hiding this comment.
Does this flow change significantly for the case where the user is working with an existing project that already contains a bridging header?
There was a problem hiding this comment.
No except that they'd not see the prompt (if there's already a bridging header file). I plan to cover in a future demo app the case with an existing project.
* [iOS][GPU] Add iOS GPU workflow (#1200) * pt mobile script and optimize recipe (#1193) * pt mobile script and optimize recipe * 1 pt mobile new recipes summary and 5 recipes * updated recipes_index.rst * thumbnail png fix for ios recipe in recipes_index.rst * edits based on feedback * Updating 1.7 branch (#1205) * Update event tracking (#1188) * Update beginner_source/audio_preprocessing_tutorial.py (#1199) * Typo in beginner_source/audio_preprocessing_tutorial.py Typo in beginner_source/audio_preprocessing_tutorial.py fron > from * update title. * fix file access. Co-authored-by: JuHyuk Park <creduo@gmail.com> * Update audio_preprocessing_tutorial.py (#1202) Adds a comment for running this tutorial in Google Colab. Co-authored-by: Pat Mellon <16585245+patmellon@users.noreply.github.com> Co-authored-by: Vincent QB <vincentqb@users.noreply.github.com> Co-authored-by: JuHyuk Park <creduo@gmail.com> Co-authored-by: Tao Xu <taox@fb.com> Co-authored-by: Jeff Tang <jeffxtang@fb.com> Co-authored-by: Pat Mellon <16585245+patmellon@users.noreply.github.com> Co-authored-by: Vincent QB <vincentqb@users.noreply.github.com> Co-authored-by: JuHyuk Park <creduo@gmail.com>
* [iOS][GPU] Add iOS GPU workflow (pytorch#1200) * pt mobile script and optimize recipe (pytorch#1193) * pt mobile script and optimize recipe * 1 pt mobile new recipes summary and 5 recipes * updated recipes_index.rst * thumbnail png fix for ios recipe in recipes_index.rst * edits based on feedback * Updating 1.7 branch (pytorch#1205) * Update event tracking (pytorch#1188) * Update beginner_source/audio_preprocessing_tutorial.py (pytorch#1199) * Typo in beginner_source/audio_preprocessing_tutorial.py Typo in beginner_source/audio_preprocessing_tutorial.py fron > from * update title. * fix file access. Co-authored-by: JuHyuk Park <creduo@gmail.com> * Update audio_preprocessing_tutorial.py (pytorch#1202) Adds a comment for running this tutorial in Google Colab. Co-authored-by: Pat Mellon <16585245+patmellon@users.noreply.github.com> Co-authored-by: Vincent QB <vincentqb@users.noreply.github.com> Co-authored-by: JuHyuk Park <creduo@gmail.com> Co-authored-by: Tao Xu <taox@fb.com> Co-authored-by: Jeff Tang <jeffxtang@fb.com> Co-authored-by: Pat Mellon <16585245+patmellon@users.noreply.github.com> Co-authored-by: Vincent QB <vincentqb@users.noreply.github.com> Co-authored-by: JuHyuk Park <creduo@gmail.com>
SCRIPT AND OPTIMIZE FOR MOBILE RECIPE for PyTorch Mobile