Making lazy predictions
The model you trained last time was good, but it could be better if you passed through the training data a few more times. Also, it is a shame to see a good model go to waste, so you should use this one to make some predictions on a separate dataset from the one you train on.
An unfitted version of the model you created in the last exercise is available in your environment as dask_model. Dask DataFrames of training data are available as dask_X and dask_y.
This exercise is part of the course
Parallel Programming with Dask in Python
Exercise instructions
- Create a for loop and use it to train
dask_modelondask_Xanddask_y5 times. - Use the fitted model to make predictions for the input variables
dask_X. - Compute these predictions using the default scheduler.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Loop over the training data 5 times
____:
dask_model.____
# Use your model to make predictions
y_pred_delayed = ____
# Compute the predictions
y_pred_computed = ____
print(y_pred_computed)