installing tensorflow-model-analysis on Google Cloud datalab

Question

I am running the followoing command:

!pip install tesnorflow-model-analysis

and receiving the following error:

Collecting tesnorflow-model-analysis
  Could not find a version that satisfies the requirement tesnorflow-model-analysis (from versions: )
No matching distribution found for tesnorflow-model-analysis

The environment on Google Cloud Datalab is the following:

!python -V
Python 2.7.15 :: Anaconda, Inc.

!pip show tensorflow
Name: tensorflow
Version: 1.11.0
Summary: TensorFlow is an open source machine learning framework for everyone.
Home-page: https://www.tensorflow.org/
Author: Google Inc......

1st question: what are the requirements to make this library working?

One I have it install, I would like to have it running on Google Cloud Datalab.

There are commands to have it available on jupyter notebooks.

jupyter nbextension enable --py widgetsnbextension
jupyter nbextension install --py --symlink tensorflow_model_analysis
jupyter nbextension enable --py tensorflow_model_analysis

2nd question: what are the equivalents on Google Datalab Notebook?

Many thanks, eilalan

Answer 1

1st question: what are the requirements to make this library working?

It looks like there is a typo in the package name; try !pip install tensorflow-model-analysis instead.

Answer 2

For what its worth, I also do my work on GCP with datalab. I was pleasantly surprised that I could get my estimator based code working in a jupyterlab notebook. On datalab I use python3. On jupyterlab I use python2 and once I got the google cloud sdk 236.0.0 installed I could even do prediction using %%bash gcloud ml-engine --local working.