EfficientNet-PyTorch

EfficientNet-PyTorch copied to clipboard

Hello is there any example training script? For example Cat vs Dogs or flower classification. Can somebody tell me how can I use this repo to train my model with...

guitar9

Reduce of accuracy after update to version 0.7

1

Hi, noticed that after upgrade to version 0.7 accuracy of trained models from 0.63 reduces on a few percentage. And it would be not so critical, but retraining from scratch...

sh1ng

RuntimeError: ONNX export failed: Couldn't export Python operator SwishJitAutoFn

2

When I exported b7 to onnx: my env is torch 1.5, opset11.and I try other's version of torch or opset, it will meet other problems. someone meet this error? Please...

DJMeng

bn momentum is different

3

Dear all, Even if navigate into this code is harder than the 10 years journey of Odisseo, I have noticed that in `_global_params` you set `batch_norm_momentum` to `0.99`, if you...

FrancescoSaverioZuppichini

Hi, I was trying to train efficientNet b2, b3, and b4 using apex mixed precision. I use efficientNet as backbone for a segmentation task using CityScapes. Unfortunately, after a few...

cuevhv

Can not load from pretrained?

3

Can not load from pretrained? ``` Traceback (most recent call last): File "/data/magiclens/apps/decensored_server/.venv/lib/python3.6/site-packages/celery/app/trace.py", line 385, in trace_task R = retval = fun(*args, **kwargs) File "/data/magiclens/apps/decensored_server/.venv/lib/python3.6/site-packages/celery/app/trace.py", line 648, in __protected_call__ return...

playma

How to train the network with my dataset? and resize the input size?

1

KrystalCWT

AttributeError: 'NoneType' object has no attribute 'missing_keys'

3

#### when I run your example it comes this: ``` File "anaconda3/envs/sikuenv/lib/python3.6/site-packages/efficientnet_pytorch/model.py", line 196, in from_pretrained load_pretrained_weights(model, model_name, load_fc=(num_classes == 1000)) File "anaconda3/envs/sikuenv/lib/python3.6/site-packages/efficientnet_pytorch/utils.py", line 298, in load_pretrained_weights assert str(res.missing_keys) ==...

gcy506067668

I made some changes based on the b0 model and i want to pretrain it on ImageNet , can you guide me how to do it?

zyx960926

How to add additional layer in pre-trained model?

3

Can you please guide me how to add some extra fully connected layer on top of a pre-trained model ``` from efficientnet_pytorch import EfficientNet model = EfficientNet.from_pretrained('efficientnet-b0') ``` I am...

talhaanwarch