Missing PIL from AzureML TF 2.3 curated environment
Tibor Sloboda
1
Reputation point
I've been attempting to run my machine learning experiments (computer vision) on the curated environment AzureML provides, that has TensorFlow 2.3 (whose dependency should be PIL), but I keep getting an error stating that PIL is not present. Specifically, it crashed at it saying that PIL.Image cannot be imported.
I have run my script on my local computer without issue using the same dependencies (as saved from the curated environment). The error only seems to happen on the remote during a run.
Sign in to answer