Skip to content
This repository has been archived by the owner on Sep 18, 2024. It is now read-only.

add pruner unit test #1771

Merged
merged 3 commits into from
Nov 25, 2019
Merged

add pruner unit test #1771

merged 3 commits into from
Nov 25, 2019

Conversation

tanglang96
Copy link
Contributor

add l1filter_pruner and slim_pruner unit test

@@ -34,6 +34,6 @@ We implemented one of the experiments in ['Learning Efficient Convolutional Netw
| Model | Error(paper/ours) | Parameters | Pruned |
| ------------- | ----------------- | ---------- | --------- |
| VGGNet | 6.34/6.40 | 20.04M | |
| Pruned-VGGNet | 6.20/6.39 | 2.03M | 88.5% |
| Pruned-VGGNet | 6.20/6.26 | 2.03M | 88.5% |
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

could you provide some settings in those experiments? like hyper-parameters.

@@ -169,7 +169,7 @@ def main():
new_model.to(device)
new_model.load_state_dict(torch.load('pruned_vgg19_cifar10.pth'))
test(new_model, device, test_loader)
# top1 = 93.61%
# top1 = 93.74%
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

does this value will change by some random seed? If will, we should not assume the top1 value is a specific value.

@chicm-ms chicm-ms merged commit 503a357 into microsoft:master Nov 25, 2019
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants