Konstantin Willeke comments

Results 8 comments of


                                            Konstantin Willeke

No error is thrown when a table name contains `_`

Thanks a lot for picking it up! Cheers from germany :-)

Improving the default model flexibility

Hi! Yes thank you for these suggestions. Yes our `Core` modules aren't completely flexible. But this would be a good functionality to add.

A couple of issues in model_predictions from submission.py

Hi, thank your for raising these issues, and we'll look into it. A quick question: Did these errors occur when you have used your own model? You are right that...

RuntimeError: No CUDA GPUs are available even when using device='cpu'

Thanks, we'll look into that one!

Improving model-training notebook and 'trainers.py'

Hi - yes I agree that (1) is an important feature. Has the validation accuracy been a NaN value? Same for (2), it'd be a great addition to make it...

Loading a pre-trained model core to get responses to images

Hi there! thank you for your questions. Yes it is possible to load the pretrained model. Have a look here: https://github.com/sinzlab/sensorium/blob/main/notebooks/model_tutorial/2_model_evaluation_and_inspection.ipynb you can just run this: ``` model.load_state_dict(torch.load("./model_checkpoints/pretrained/generalization_model.pth")); model.cuda().eval(); #...

Loading a pre-trained model core to get responses to images

Oh yes, by default all parts of the model and dataloaders are transferred to a GPU. But you can disable it by adding `"cuda": False` to the dataset_config. so this...

Loading a pre-trained model core to get responses to images

I'm sorry for all of these issues! Your dataloader is missing some data_keys. Could it be that you only added one dataset to the dataset config, and not all of...