pytorch-retraining
Transfer Learning shootout for PyTorch's model zoo (torchvision).
model_pretrained, diff = load_model_merged('inception_v3', num_classes)
final_param_names = [d[0] for d in diff]stats = train_eval(model_pretrained, trainloader, testloader, final_params_names)
Chart training_time
, evaluation_time
(fps), top-1 accuracy
for varying levels of retraining depth (shallow, deep and from scratch)
| | |:---:| | Transfer learning on example dataset Bee vs Ants with 2xK80 GPUs|
Results on more elaborate Dataset
num_classes = 23, slightly unbalanced, high variance in rotation and motion blur artifacts with 1xGTX1080Ti
| | |:---:| | Constant LR with momentum |
| | |:---:| | Cyclical Learning Rate |