Applied Deep Learning with Python

Activity:Optimizing a Deep Learning Model

In this activity, we implement different optimization strategies to the model created in Chapter 5, Model Architecture (bitcoin_lstm_v0). That model achieves a MAPE performance on the complete denormalization test set of about 8.4 percent. We will try to reduce that gap.

1. Using your terminal, start a TensorBoard instance by executing the following command:

      $ cd ./chapter_3/activity_7/      $ tensorboard --logdir=logs/ 

2. Open the URL that appears on the screen and leave that browser tab open, as well. Also, start a Jupyter Notebook instance with:

       $ jupyter notebook

Open the URL that appears in a different browser window.

3. Now, open the Jupyter Notebook called Activity_7_Optimizing_a_deep_ learning_model.ipynb ...