Update Deploy Seq2Seq Tutorial with New TorchScript API #565
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Deploy preview for pytorch-tutorials-preview ready! Built with commit 9fa7632 https://deploy-preview-565--pytorch-tutorials-preview.netlify.com |
Chillee
reviewed
Jul 19, 2019
| # training. | ||
| # | ||
| # What is the Hybrid Frontend? | ||
| # What is the TorchScript? |
| # :align: center | ||
| # :alt: workflow | ||
| # control flow, a **scripting** mechanism is provided. The | ||
| # ``torch.jit.script`` function takes module or function and does not |
Contributor
There was a problem hiding this comment.
The torch.jit.script decorator takes a module or function...
Contributor
Author
There was a problem hiding this comment.
torch.jit.script right now does not necessarily to be a decorator right?
Contributor
Author
There was a problem hiding this comment.
like you can call it scripted_model = torch.jit.script(model), which is not a decorator.
| # ``torch.jit.script`` function takes module or function and does not | ||
| # requires example inputs. Scripting then explicitly converts the module | ||
| # or function code to TorchScript, including all possible control flow | ||
| # routes. The one caveat with using scripting is that it only supports |
Contributor
There was a problem hiding this comment.
probably "all control flow" instead of "all possible control flow routes"?
Contributor
There was a problem hiding this comment.
Also, I think "One caveat with using scripting" is a bit more natural than "The one caveat..."
| # ~~~~~~~~~~~~~~~~~~~~~~ | ||
| # | ||
| # Similarly to the ``EncoderRNN``, this module does not contain any | ||
| # Similarly to the ``EncoderRNN```, this module does not contain any |
Contributor
There was a problem hiding this comment.
Is the extra ` on purpose?
wanchaol
force-pushed
the
seq2seq
branch
2 times, most recently
from
ecbeae4 to
88bda68
Compare
Chillee
reviewed
Jul 23, 2019
| # -*- coding: utf-8 -*- | ||
| """ | ||
| Deploying a Seq2Seq Model with the Hybrid Frontend | ||
| Deploying a Seq2Seq Model with the TorchScript |
Member
|
This looks good to me, but needs a rebase @wanchaol |
wanchaol
force-pushed
the
seq2seq
branch
3 times, most recently
from
a2e366c to
f5bce02
Compare
* Update requirements.txt Pinning sphinx gallery. Step one. Please don't merge. * Update build.sh removed audio, audio sample, added || true to line 39. * Update build.sh uninstall/install awscli * Update build.sh Trying to unpin Sphinx from ==1.8.2. * Update build.sh Pinning sphinx-gallery. * Update build.sh repining sphinx trying pip install * Update build.sh removed -y option * Update build.sh * Update build.sh
Updated the Hybrid frontend section to link to JIT/Torchscript.
Some spelling errors fixed.. "That looks waaay better than chance" "Because your network is realllly small"
Corrected to CHW not HWC (3x36x36)
Updated some code to remove eval in the code.
Removed old ONNX tutorial from TOC.
Added torchaudio from source.
Missed a #.
Edited sentence.
updating for no version number
If you don't configure this string of code, you will get an error when you iterate over the update from 4000_checkpoint.tar: ``` encoder_optimizer.step() ``` Error message: ``` exp_avg.mul_(beta1).add_(1 - beta1, grad) RuntimeError: expected backend CPU and dtype Float but got backend CUDA and dtype Float ``` Fix it: pytorch/pytorch#2830 ``` with torch.no_grad(): correct = 0 total = 0 for images, labels in test_loader: images = images.to(device) # missing line from original code labels = labels.to(device) # missing line from original code images = images.reshape(-1, 28 * 28) out = model(images) _, predicted = torch.max(out.data, 1) total += labels.size(0) correct += (predicted == labels).sum().item() ```
If you don't configure this string of code, you will get an error when you iterate over the update from 4000_checkpoint.tar: ``` encoder_optimizer.step() ``` Error message: ``` exp_avg.mul_(beta1).add_(1 - beta1, grad) RuntimeError: expected backend CPU and dtype Float but got backend CUDA and dtype Float ``` Fix it: pytorch/pytorch#2830 ``` model = Model() model.load_state_dict(checkpoint['model']) model.cuda() optimizer = optim.Adam(model.parameters()) optimizer.load_state_dict(checkpoint['optimizer']) for state in optimizer.state.values(): for k, v in state.items(): if isinstance(v, torch.Tensor): state[k] = v.cuda() ```
Tensorboard dependency.
Contributor
Author
|
close this in favor of #597 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
12 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.