Add weight for mnasnet0_75 and mnasnet1_3 by YosuaMichael · Pull Request #6019 · pytorch/vision (original) (raw)
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service andprivacy statement. We’ll occasionally send you account related emails.
Already on GitHub?Sign in to your account
Conversation5 Commits6 Checks0 Files changed
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
[ Show hidden characters]({{ revealButtonHref }})
Adding mnasnet0_75 and mnasnet1_3 weights.
Training commands:
# Training mnas0_75
python -u ~/script/run_with_submitit.py \
--timeout 6000 --ngpus 8 --nodes 1 --batch-size=128 \
--partition train --model mnasnet0_75 \
--data-path="/datasets01_ontap/imagenet_full_size/061417" \
--lr=0.5 --lr-scheduler=cosineannealinglr --lr-warmup-epochs=5 --lr-warmup-method=linear \
--auto-augment=ta_wide --epochs=600 --random-erase=0.1 --weight-decay=0.00002 \
--norm-weight-decay=0.0 --label-smoothing=0.1 --mixup-alpha=0.2 --cutmix-alpha=1.0 \
--train-crop-size=176 --model-ema --val-resize-size=232 --ra-sampler --ra-reps=4
# Training mnas1_3
python -u ~/script/run_with_submitit.py \
--timeout 6000 --ngpus 8 --nodes 1 --batch-size=128 \
--partition train --model mnasnet1_3 \
--data-path="/datasets01_ontap/imagenet_full_size/061417" \
--lr=0.5 --lr-scheduler=cosineannealinglr --lr-warmup-epochs=5 --lr-warmup-method=linear \
--auto-augment=ta_wide --epochs=600 --random-erase=0.1 --weight-decay=0.00002 \
--norm-weight-decay=0.0 --label-smoothing=0.1 --mixup-alpha=0.2 --cutmix-alpha=1.0 \
--train-crop-size=176 --model-ema --val-resize-size=232 --ra-sampler --ra-reps=4
Validation script:
# validate mnasnet0_75
python -u ~/script/run_with_submitit.py \
--timeout 3000 --nodes 1 --ngpus 1 --batch-size=1 \
--partition train --model mnasnet0_75 \
--data-path="/datasets01_ontap/imagenet_full_size/061417" \
--weights="MNASNet0_75_Weights.IMAGENET1K_V1" \
--test-only \
# Acc@1 71.180 Acc@5 90.496
# validate mnasnet1_3
python -u ~/script/run_with_submitit.py \
--timeout 3000 --nodes 1 --ngpus 1 --batch-size=1 \
--partition train --model mnasnet1_3 \
--data-path="/datasets01_ontap/imagenet_full_size/061417" \
--weights="MNASNet1_3_Weights.IMAGENET1K_V1" \
--test-only \
# Acc@1 76.506 Acc@5 93.522
YosuaMichael changed the title
Add weight for mnasnet0_75 and mnasnet1_3 [WIP] Add weight for mnasnet0_75 and mnasnet1_3
datumbox changed the title
[WIP] Add weight for mnasnet0_75 and mnasnet1_3 Add weight for mnasnet0_75 and mnasnet1_3
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
You forgot to change the default values for the weights in handle_legacy_interface
. Other than that it looks good.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The failed test seems to not related to this PR, will merge.
facebook-github-bot pushed a commit that referenced this pull request
Summary:
Add weight for mnasnet0_75 and mnasnet1_3
Fix missing comma
Add PR url as recipe, and update the metrics
Add weights to legacy handler
Update docs to specify there are weights available
Reviewed By: NicolasHug
Differential Revision: D36760931
fbshipit-source-id: 00211a6dd22b4b42a9845b7b4d25337ed14a6349