I am trying to work in tensorflow in both python (via PyCharm) to train the model, and Jupyter notebook to analyze the results.
However, after I restore a session checkpoint & analyze the data in jupyer, I cant retrain a model at the same time due to the " Blas GEMM launch failed" error. I know this issue can be resolved by restarting the jupyter kernel (see here), however I would prefer not having to do that. Is there any way of switching back and forth between the two without have to restart the ipython kernel?
Thanks!