I created a pipeline for classification problem in Colab Notebook with pytorch version 1.7.0+cu101 and the model is converging very well there. But if I use the same pipeline to execute in my local laptop, the accuracy never goes above 10% with pytorch version 1.7.1
I have no clue, what am I missing here. Please help.
If you are using different operating systems (Linux in Colab vs. Windows on your Laptop), make sure the data paths are equal, as e.g. the sorting is not equal on those systems.
If that doesn’t help, try to update Colab to 1.7.1 and check, if you are still able to train the model or if there might be a potential issue in the latest PyTorch release.