[FIX] ResNet-18 accuracy 68.934 -> 69.752 #62
Merged
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Outdated suggestions cannot be applied.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.
The problem was with resize step in the preprocessing.
The problem is pretty much described in resize docs: "The output image might be different depending on its type: when downsampling, the interpolation of PIL images and tensors is slightly different, because PIL applies antialiasing. This may lead to significant differences in the performance of a network".
For ResNet-18, accuracy still slightly differs from https://pytorch.org/vision/stable/models.html. Our preprocessing pipeline (without PIL) achieves 69.752, when pytroch pipeline (with PIL) achieves 69.758.