training-tensorflow-model.md

Training Tensorflow Model

Introduction

TensorFlow is an open-source machine learning software library, TensorFlow is used to train neural networks. Expressed in the form of stateful dataflow graphs, each node in the graph represents the operations performed by neural networks on multi-dimensional arrays. These multi-dimensional arrays are commonly known as “tensors”, hence the name TensorFlow. In this example, we will be training a MNIST model.

Training TensorFlow models Locally​

This section is from TensorFlow 2 quickstart for beginners

TensorFlow 2 quickstart for beginners​

This short introduction uses Keras to:

1. Load a prebuilt dataset.
2. Build a neural network machine learning model that classifies images.
3. Train this neural network.
4. Evaluate the accuracy of the model.

Set up TensorFlow​

Import TensorFlow into your program to check whether it is installed

import tensorflow as tf
import os
print("TensorFlow version:", tf.__version__)

mkdir /inputs
wget https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz -O /inputs/mnist.npz

mnist = tf.keras.datasets.mnist

CWD = '' if os.getcwd() == '/' else os.getcwd()
(x_train, y_train), (x_test, y_test) = mnist.load_data('/inputs/mnist.npz')
x_train, x_test = x_train / 255.0, x_test / 255.0

Build a machine-learning model​

Build a tf.keras.Sequential model by stacking layers.

model = tf.keras.models.Sequential([
  tf.keras.layers.Flatten(input_shape=(28, 28)),
  tf.keras.layers.Dense(128, activation='relu'),
  tf.keras.layers.Dropout(0.2),
  tf.keras.layers.Dense(10)
])

For each example, the model returns a vector of logits or log-odds scores, one for each class.

predictions = model(x_train[:1]).numpy()
predictions

The tf.nn.softmax function converts these logits to probabilities for each class:

tf.nn.softmax(predictions).numpy()

Note: It is possible to bake the tf.nn.softmax function into the activation function for the last layer of the network. While this can make the model output more directly interpretable, this approach is discouraged as it's impossible to provide an exact and numerically stable loss calculation for all models when using a softmax output.

Define a loss function for training using losses.SparseCategoricalCrossentropy, which takes a vector of logits and a True index and returns a scalar loss for each example.

loss_fn = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)

This loss is equal to the negative log probability of the true class: The loss is zero if the model is sure of the correct class.

This untrained model gives probabilities close to random (1/10 for each class), so the initial loss should be close to -tf.math.log(1/10) ~= 2.3.

loss_fn(y_train[:1], predictions).numpy()

Before you start training, configure and compile the model using Keras Model.compile. Set the optimizer class to adam, set the loss to the loss_fn function you defined earlier, and specify a metric to be evaluated for the model by setting the metrics parameter to accuracy.

model.compile(optimizer='adam',
              loss=loss_fn,
              metrics=['accuracy'])

Train and evaluate your model​

Use the Model.fit method to adjust your model parameters and minimize the loss:

model.fit(x_train, y_train, epochs=5)

The Model.evaluate method checks the models performance, usually on a "Validation-set" or "Test-set".

model.evaluate(x_test,  y_test, verbose=2)

The image classifier is now trained to ~98% accuracy on this dataset. To learn more, read the TensorFlow tutorials.

If you want your model to return a probability, you can wrap the trained model, and attach the softmax to it:

probability_model = tf.keras.Sequential([
  model,
  tf.keras.layers.Softmax()
])

probability_model(x_test[:5])

mkdir /outputs

The following method can be used to save the model as a checkpoint

model.save_weights('/outputs/checkpoints/my_checkpoint')

ls /outputs/

Running on Bacalhau​

The dataset and the script are mounted to the TensorFlow container using an URL, we then run the script inside the container

Declarative job description​

The same job can be presented in the declarative format. In this case, the description will look like this:

name: Training ML model using tensorflow
type: batch
count: 1
tasks:
  - name: My main task
    Engine:
      type: docker
      params:
        WorkingDirectory: "/inputs"
        Image: "tensorflow/tensorflow" 
        Entrypoint:
          - /bin/bash
        Parameters:
          - -c
          - python train.py
    InputSources:
      - Source:
          Type: urlDownload
          Params:
            URL: https://storage.googleapis.com/tensorflow/tf-keras-datasets/mnist.npz
        Target: /inputs
      - Source:
          Type: urlDownload
          Params:
            URL: https://gist.githubusercontent.com/js-ts/e7d32c7d19ffde7811c683d4fcb1a219/raw/ff44ac5b157d231f464f4d43ce0e05bccb4c1d7b/train.py
        Target: /inputs
    Resources:
      GPU: "1"

The job description should be saved in .yaml format, e.g. tensorflow.yaml, and then run with the command:

bacalhau job run tensorflow.yaml

Checking the State of your Jobs​

Job status​

You can check the status of the job using bacalhau job list.

bacalhau job list --id-filter ${JOB_ID}

When it says Completed, that means the job is done, and we can get the results.

Job information​

You can find out more information about your job by using bacalhau job describe.

bacalhau job describe ${JOB_ID}

Job download​

You can download your job results directly by using bacalhau job get. Alternatively, you can choose to create a directory to store your results. In the command below, we created a directory and downloaded our job output to be stored in that directory.

rm -rf results && mkdir -p results
bacalhau job get $JOB_ID --output-dir results

After the download has finished you should see the following contents in results directory

Viewing your Job Output​

Now you can find the file in the results/outputs folder. To view it, run the following command:

cat results/outputs/

Support​

If you have questions or need support or guidance, please reach out to the Bacalhau team via Slack (#general channel).