Issue with time series forecasting with pycaret notebook

[marijaselakovic]

When running automl_timeseries_forecasting_with_pycaret.ipynb notebook the mlflow-cratedb module gets installed but it is not found when importing:

ModuleNotFoundError                       Traceback (most recent call last)
[<ipython-input-3-c9198aa96905>](https://localhost:8080/#) in <cell line: 6>()
      4 import plotly
      5 import plotly.graph_objects as go
----> 6 import mlflow_cratedb  # Required to enable the CrateDB MLflow adapter.
      7 from dotenv import load_dotenv
      8 

ModuleNotFoundError: No module named 'mlflow_cratedb'

Can you check/test the code? The issue persists both locally and in Colab.

[Andreas Motl]

CI says it works, so the flaw might be on your local environment and/or on Colab.

Maybe some error can be observed while installing the packages? To check what might go wrong, just create a separate throw-away virtualenv and probe if the installation works, or channel back any error messages. Thanks!

python3 -m venv .venv
source .venv/bin/activate
pip install mlflow-cratedb
python

>>> import mlflow_cratedb

[marijaselakovic]

I created a fresh virtual environment, but the only solution I found so far is installing the package with %pip install mlflow_cratedb command. For some reason, the package wasn't visible in the notebook.