dfencoder
dfencoder copied to clipboard
AlliedToasters
The classifier model does not work at the moment, as "target_df" doesn't contain "label_col" column (removed in self.prepare_df). Did anyone solve this? _Originally posted by @IdoCohenAB in https://github.com/AlliedToasters/dfencoder/issues/19#issuecomment-1459991219_
I am trying to figure out how to utilize multiple GPUs for training
I couldn't find a function for getting swapped, and unaltered loss for a df using an already trained model. I can create a new PR to add a function like...
Hi Michael, I started using your great lib dfencoder. I am having issues with running demo. As you advised, created a virtual environment and running the demo code. I tried...
The HLD is helpful, but what I think is still hard to understand is how you actually train the model, speak how inputs and outputs of the model look like...
Hello, when I ran the code on my computer, I reported the following error **model = AutoEncoder( encoder_layers = [512, 512, 512], #model architecture decoder_layers = [], #decoder optional -...
Received the following error in cell 6 of the demo.ipynb: --------------------------------------------------------------------------- RuntimeError Traceback (most recent call last) in ----> 1 model.fit(X_train, epochs=1000, val=X_val) C:\Program Files\Anaconda3\envs\my-fastai2019\lib\site-packages\dfencoder\autoencoder.py in fit(self, df, epochs, val)...
Hi, Thanks you for the very interesting library! Looking at the code I was wondering how you were balancing the different losses scales. The MSE loss is probably on a...
dfencoder 0.0.42 torch version: 1.12.0+cu102 dill version: 0.3.4 Error: Can't pickle : it's not found as torch._VariableFUnctionsClass.relu. Any suggestion on how to solve it?
