# TensorFlow

## In this learning path

- [Introduction](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/rpi/)
- [Introduction to the Raspberry Pi 4](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/rpi/intro/)
- [Setup a Raspberry Pi 4 and an Arm cloud instance](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/rpi/setup/)
- [Identifying the hardware](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/rpi/id/)
- [Linux Kernel Compile](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/rpi/kernel/)
- [TensorFlow](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/rpi/tf/)
- [Docker](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/rpi/docker/)
- [Linux Perf](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/rpi/perf/)
- [Next Steps](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/rpi/_next-steps/)

## TensorFlow for machine learning
To compare the performance of TensorFlow on the Raspberry Pi 4 and the Arm cloud server, install it and run an example.

Install `pip` and `python3`.

```bash
sudo apt install python-is-python3 python3-pip -y
```

Install TensorFlow.

```bash
pip install tensorflow-aarch64 tensorflow_io
```

You can now follow the instructions in the [TensorFlow quickstart example](https://www.tensorflow.org/tutorials/quickstart/beginner) or proceed to the steps in the Quickstart example below.

## Quickstart example
To save time entering the commands from the TensorFlow example, the code is shared here.

Using a text editor of your choice, copy the contents below and save it in a file named `example.py`:

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

mnist = tf.keras.datasets.mnist

(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0

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)
])

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

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

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

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

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

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

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

probability_model(x_test[:5])
```

Run the example using `python`.

```bash
time python ./example.py
```

## Results
The results below show the time taken to run the tensorflow example on the Raspberry Pi and the Arm server. This gives you an idea of the relative performance.

| System             | Time to complete |
|--------------------|------------------|
| Raspberry Pi 4     | 1 min 46 sec     |
| Arm server (4 vCPU) | 22 sec           |