README.md

Downstream tasks training scripts

This folder contains reference training scripts for Faster/Mask/Keypoint-RCNN-ResNet50-FPN for object detection, segmentation and keypoint detection.

To execute the example commands below you must install the following:

cython
pycocotools
matplotlib

You must modify the following flags:

--data-path=/path/to/coco/dataset

--nproc_per_node=<number_of_gpus_available>

Weights backbone is adopted from the classification task, therefore it should be the compressed resnet50 with coressponding pruning ratio.

As recommended by torchvision, default learning rate and batch size values go along with 8xV100. Please modify them to match with your numbers of gpus, e.g., --nproc_per_node=1 --lr 0.02 -b 2.

Faster R-CNN ResNet-50 FPN

torchrun --nproc_per_node=8 train.py --dataset coco --model fasterrcnn_resnet50_fpn --epochs 26 --lr-steps 16 22 --aspect-ratio-group-factor 3 --weights-backbone resnet50.pt -cpr [0.]+[0.1]*3+[0.35]*16

Mask R-CNN

torchrun --nproc_per_node=8 train.py --dataset coco --model maskrcnn_resnet50_fpn --epochs 26 --lr-steps 16 22 --aspect-ratio-group-factor 3 --weights-backbone resnet50.pt -cpr [0.]+[0.1]*3+[0.35]*16

Keypoint R-CNN

torchrun --nproc_per_node=8 train.py --dataset coco_kp --model keypointrcnn_resnet50_fpn --epochs 46 --lr-steps 36 43 --aspect-ratio-group-factor 3 --weights-backbone resnet50.pt -cpr [0.]+[0.1]*3+[0.35]*16

Visualizing model inference

Compressed models can be deployed as torchvision's guide.

For your convenience, we have also prepared an example script, visualize.py, that emphasizes the enhanced FPS achieved by the pruned model. Below is the example usage:

python visualize.py --input birthday.mp4 --custom -cpr [0.]+[0.1]*3+[0.35]*16 --weight model_24.pth

By using this script, you can effortlessly visualize and compare the inference speed of both the baseline and pruned models. This provides a clear demonstration of the substantial throughput acceleration achieved by CORING's compression techniques.