Training and saving a TensorFlow model
To deploy a deep learning model in production, you need to train it. The Python script
- Loads the MNIST dataset containing handwritten digits and their labels
- Defines a small Multilayer Perceptron in TensorFlow
- Trains this model on MNIST data and stores the resulting model to
You can click on
train.py in the editor window for details. Or you can run this script by clicking here:
model.pb appear in the file tree of the editor window as the script runs.
With that, we can deploy the TensorFlow model with the Skymind Intelligence Layer (SKIL).