The following guide are based on the tutorials from https://www.tensorflow.org/tutorials.
- Python 3 (of course)
- Tensorflow 2.0 (of course)
Get your environment up and running. Should be a piece of cake if you have followed https://github.com/acntech/vagrant-ml-developer...
Run the guides in a Jupyter notebook (in terminal: jupyter notebook
).
If you are not very experienced in Tensorflow and Keras, I recommend you to start with tf2_in_a_nutshell.ipynb
. Run through the notebook and do the suggested assignments at the bottom of the script.
If you have some experience with both Tensorflow and Keras, you might want to skip directly to tf2_advanced.ipynb
. Run through the notebook and do the suggested assignments at the bottom of the script.
Not enough? Or, tired of MNIST basics? Select one of the great tutorials on https://www.tensorflow.org/tutorials.
Some of my recommendations grouped by problem type (you might want to run some of these in Google Colab because of training time):
- Structured data: https://www.tensorflow.org/tutorials/structured_data/feature_columns
- Text classification: https://www.tensorflow.org/tutorials/text/text_classification_rnn
- Time series: https://www.tensorflow.org/tutorials/structured_data/time_series
- GAN: https://www.tensorflow.org/tutorials/generative/dcgan