ML_Decoder

ML_Decoder copied to clipboard

What should I do if I want to train ML_Decoder on a custom dataset. [multi-label task]

Q1ngS0ng

Training commands on CIFAR100

1

What are the specific training parameters for the training commands on CIFAR100？

xiaohua-chen

here is my code `model = create_model_test(args, load_head=True).cuda() ########### eliminate BN for faster inference ########### model.load_state_dict(torch.load(args.model_path, map_location='cpu')) model = model.cpu() model = InplacABN_to_ABN(model) model = fuse_bn_recursively(model) model = model.cpu() `...

eguoguo321

When loading a model with `load_head=True` it would throw an exception because the `key` variable isn't defined. This PR moves the variable definition to fix this bug.

abhaybd

hi, thanks for your wonderful work. Could you please provide the NUS-WIDE weights just as you provided for MSCOCO and other datasets?

aliman80

Hi, thanks for your excellent work. I am trying to apply it on my custom dataset with 4 classes with 2 seen and 2 unseen classes . I m tryng...

aliman80

when i load model, not found key variable

1

https://github.com/Alibaba-MIIL/ML_Decoder/blob/8a9e984f671c9c30c98d2c45dfcaf4383381c254/src_files/models/utils/factory.py#L47-L56 when load_head is True, it didn't defined `key`. \+ i know solution. so, i just notified this problem. plz fix it

JadeKim042386