I want to process a large dataset with deep learning algorithms, which were implemented in the TensorFlow framework. I used Dask for loading the data, but the model’s accuracy reached 100% in the second epoch. In the same manner, I used Numpy for data loading in the system with higher RAM, and the results didn’t achieve 100% even after 50 epochs.
How can I fix the problem in Dask?
New contributor
Mbahrami is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.