Hints

Hint #1

Click to reveal Hint #1

  • Maybe installing some missing libraries with !pip

  • Some of the parameters in the code could be wrong…​

Hint #2

Click to reveal Hint #1

GH link: https://github.com/rh-aiservices-bu/ai-mazing-race

git clone https://github.com/rh-aiservices-bu/ai-mazing-race.git

  • Create a project (stars-userX)

  • small - pytorch - wb, with 5 GB of disk.

  • Navigate to lab-materials/03 and open stars-classification.ipynb

  • train_test_split has something wrong…​ maybe changing some range in the parameters?

  • plt.show() is used to visualize the confusion matrix

  • In the Jupyter Notebook, you have the possibility to use Elyra pipelines…​

  • And then from the Data Science Pipelines UI you can setup pipeline schedule.

Hint #3 (Complete solution)

Click to reveal Hint #3

  • !pip install seaborn

  • adjust the test_size to 0.3 from train_test_split in "Train the Model" section as the 30% of the data is used for testing

  • substitute plt.show_me_the_stars with plt.show()

  • From the Jupyter Notebook add a New Data Science Pipeline Editor, and drag and drop the stars-classification.ipynb file

  • Run the pipeline from Elyra / Data Science Pipeline Editor

  • Open the OpenShift AI dashboard, go to Data Science Pipelines and setup schedule for 24 hours