Chapter 4 - Replacing the Classifier

[ldkong1205]

transfer_model.fc = nn.Sequential(nn.Linear(transfer_model.fc.in_features, 500),
nn.ReLU(),
nn.Dropout(),
nn.Linear(500,2))

AttributeError: 'Sequential' object has no attribute 'in_features'

[MarcusFra]

Hmm, I cannot reproduce this error.
Did you run the https://github.com/falloutdurham/beginners-pytorch-deep-learning/blob/master/chapter4/Chapter%204.ipynb or did you copy the code from the book?
Maybe it's due to the defintion of transfer_model - did you run transfer_model = models.resnet50(pretrained=True) earlier?
After loading this pretrained model you can check the structure of the fc block by print(transfer_model.fc). It should return Linear(in_features=2048, out_features=1000, bias=True).

After redefining this with transfer_model.fc = nn.Sequential(nn.Linear(transfer_model.fc.in_features, 500), ...)) you should get

Sequential(
  (0): Linear(in_features=2048, out_features=500, bias=True)
  (1): ReLU()
  (2): Dropout(p=0.5, inplace=False)
  (3): Linear(in_features=500, out_features=2, bias=True)

when you run print(transfer_model.fc).

[ldkong1205]

Hmm, I cannot reproduce this error.
Did you run the https://github.com/falloutdurham/beginners-pytorch-deep-learning/blob/master/chapter4/Chapter%204.ipynb or did you copy the code from the book?
Maybe it's due to the defintion of transfer_model - did you run transfer_model = models.resnet50(pretrained=True) earlier?
After loading this pretrained model you can check the structure of the fc block by print(transfer_model.fc). It should return Linear(in_features=2048, out_features=1000, bias=True).

After redefining this with transfer_model.fc = nn.Sequential(nn.Linear(transfer_model.fc.in_features, 500), ...)) you should get

Sequential(
  (0): Linear(in_features=2048, out_features=500, bias=True)
  (1): ReLU()
  (2): Dropout(p=0.5, inplace=False)
  (3): Linear(in_features=500, out_features=2, bias=True)

when you run print(transfer_model.fc).

Thank you for your reply. 😁