David Cavazos

What command are you using to run the pipeline? If you run the sample without any modifications, do you still get that error? We're using `save_main_session=True` explicitly when creating the...

It looks like pickling the session failed. Try importing the aiplatform modules locally in the functions that are used instead of importing it globally. I suspect maybe some update made...

Can you try removing the global imports for `aiplatform`? I suspect having those modules loaded as part of the main session is causing issues when pickling it. https://github.com/GoogleCloudPlatform/python-docs-samples/blob/d58ee70c1c6afee0c07a1c4002aaf8e9a7ec99ba/people-and-planet-ai/image-classification/train_model.py#L26-L27

Another thing to try is using `--pickle_library=cloudpickle`, it might help pickling things that the normal Python pickle cannot handle.

It looks like both `apache-beam` and `google-cloud-aiplatform` depend on `google-cloud-bigquery

It looks like all tests are passing except for BigQuery. One of the samples has a comment saying that pyarrow has to be

Unfortunately Apache Beam does not officially support Python 3.10 yet :( Although it should technically work, the Python version from which the worker boot files get pulled would be from...

The latest `apache-beam` version supports BigQuery

As per the docs, TensorFlow 2.10.0 is still using CUDA 11.2 with cuDNN 8. So I'm setting all samples to those versions. https://www.tensorflow.org/install/source#gpu

Is there any progress on this? I'm also getting negative indices in my source maps.