search

PingoLH / FCHarDNet

Fully Convolutional HarDNet for Segmentation in Pytorch
MIT License
195 stars 52 forks source link

How was weights/hardnet_petite_base.pth trained? #36

Open blake-varden opened 4 years ago

blake-varden commented 4 years ago

Hi, I'm looking to convert this repo to Tensorflow. I believe most of the operations here are standard and it should be relatively straightforward to convert to Tensorflow.

I will however need to regenerate the pretrained weights for hardnet_petite_base. I was wondering if you could share the details of how you trained this?

Is the structure, just the decoder portion of the network followed by an FC layer to the number of classes and then Trained on ImageNet for 100 epochs?

PingoLH commented 4 years ago

Hi, yes we trained it with ImageNet for the first 100 epochs out of a totally 150 epochs of cosine learning rate schedule.

blake-varden commented 4 years ago

Hi thank you. And to confirm the architecture of the petitie base is : the decoder + fc layer + softmax to classify the classes?

PingoLH commented 4 years ago

Yes, just add a fc layer with a cross entropy loss should be fine.

DonghweeYoon commented 4 years ago

How much hours have you spent for training the model with ImageNet?