3. Quick start for beginners

This section provides more detail on the steps described in the Quick start for experts section.

Complete any necessary setup to use your IPU system (see Section 1.1, IPU systems) before the following steps.

3.1. Enable the Poplar SDK

Note

It is best if you use the latest version of the Poplar SDK.

On some systems you must explicitly enable the Poplar SDK before you can use PyTorch or TensorFlow for the IPU, or the Poplar Graph Programming Framework. On other systems, the SDK is enabled as part of the login process.

Table 3.1 defines whether you have to explicitly enable the SDK and where to find it.

Table 3.1 Systems that need the Poplar SDK to be enabled and the SDK location

System

Enable SDK?

SDK location

Pod system

Yes

The SDK is in the directory where you extracted the SDK tarball.

Graphcloud

Yes

/opt/gc/poplar_sdk-ubuntu_18_04-[poplar_ver]+[build]

where [poplar_ver] is the software version number of the Poplar SDK and [build] is the build information.

Gcore Cloud

No

The SDK has been enabled as part of the login process.

To enable the Poplar SDK:

For SDK versions 2.6 and later, there is a single enable script that determines whether you are using Bash or Zsh and runs the appropriate scripts to enable both Poplar and PopTorch/PopART.

Source the single script as follows:

$ source [path_to_SDK]/enable

where [path_to_SDK] is the location of the Poplar SDK on your system.

Note

You must source the Poplar enable script for each new shell. You can add this source command to your .bashrc (or .zshrc for SDK versions later than 2.6) to do this on a more permanent basis.

If you attempt to run any Poplar software without having first sourced this script, you will get an error from the C++ compiler similar to the following (the exact message will depend on your code):

fatal error: 'poplar/Engine.hpp' file not found

If you try to source the script after it has already been sourced, then you will get an error similar to:

ERROR: A Poplar SDK has already been enabled.
Path of enabled Poplar SDK: /opt/gc/poplar_sdk-ubuntu_20_04-3.2.0-7cd8ade3cd/poplar-ubuntu_20_04-3.2.0-7cd8ade3cd
If this is not wanted then please start a new shell.

You can verify that Poplar has been successfully set up by running:

$ popc --version

This will display the version of the installed software.

3.2. Create and enable a Python virtual environment

It is good practice to work in a different Python virtual environment for each framework or even for each application. This section describes how you create and activate a Python virtual environment.

Note

You must activate the Python virtual environment before you can start using it.

The virtual environment must be created for the Python version you will be using. This cannot be changed after creation. Create a new Python virtual environment with:

$ virtualenv -p python3 [venv_name]

where [venv_name] is the location of the virtual environment.

Note

Make sure that the version of Python that is installed is compatible with the version of the Poplar SDK that you are using. See Supported tools in the Poplar SDK release notes for information about the supported operating systems and versions of tools.

To start using a virtual environment, activate it with:

$ source [venv_name]/bin/activate

where [venv_name] is the location of the virtual environment.

Now all subsequent installations will be local to that virtual environment.

3.3. Install the TensorFlow 2 wheels and validate

In order to run applications in TensorFlow 2 on an IPU, you have to install Python wheel files for the Graphcore ports of TensorFlow 2 and Keras and also for TensorFlow 2 add-ons.

3.3.1. TensorFlow 2 wheel

There are two TensorFlow 2 wheels included in the Poplar SDK, one for AMD processors and one for Intel processors. Check which processor is used on your system by running:

$ lscpu | grep name

The wheel file has a name of the form:

tensorflow-[ver]+[platform].whl

where [ver] is the version of the Graphcore port of TensorFlow 2 and [platform] defines the server details (processor and operating system) for the TensorFlow build. An example of the TensorFlow 2 wheel file for an AMD processor for Poplar SDK 3.0 is:

tensorflow-2.6.3+gc3.0.0+236842+d084e493702+amd_znver1-cp38-cp38-linux_x86_64.whl

Install the Graphcore TensorFlow 2 distribution for an AMD processor with:

$ python -m pip install ${POPLAR_SDK_ENABLED?}/../tensorflow-2.*+amd_*.whl

Install the Graphcore TensorFlow 2 distribution for an Intel processor with:

$ python -m pip install ${POPLAR_SDK_ENABLED?}/../tensorflow-2.*+intel_*.whl

POPLAR_SDK_ENABLED is the location of the Poplar SDK defined when the SDK was enabled. The ? ensures that an error message is displayed if Poplar has not been enabled.

3.3.2. Keras wheel

In the TensorFlow 2.6 release, Keras was moved into a separate pip package. In the Poplar SDK 2.6 release, which includes the Graphcore distribution of TensorFlow 2.6, there is a Graphcore distribution of Keras which includes IPU-specific extensions.

Note

The Keras wheel must be installed after the TensorFlow wheel, but before the TensorFlow Addons wheel.

The Keras wheel file has a name of the form:

keras-[tf-ver]*.whl

where [tf-ver] is the TensorFlow 2 version. An example of the Keras wheel file for TensorFlow 2.6 for the IPU for Poplar SDK 3.0 is:

keras-2.6.0+gc3.0.0+236851+1744557f-py2.py3-none-any.whl

Install the Keras wheel using the following command:

$ python -m pip install --force-reinstall --no-deps ${POPLAR_SDK_ENABLED?}/../keras-2.*.whl

POPLAR_SDK_ENABLED is the location of the Poplar SDK defined when the SDK was enabled. The ? ensures that an error message is displayed if Poplar has not been enabled.

3.3.3. TensorFlow 2 Addons wheel

IPU TensorFlow Addons is a collection of add-ons created for the Graphcore port of TensorFlow. These include layers and optimizers for Keras, as well as legacy TensorFlow layers. For more information, refer to the section on IPU TensorFlow Addons in the TensorFlow 2 user guide.

Note

  • The IPU TensorFlow 2 Addons wheel file is only available in Poplar SDK 2.4 and later.

  • There are separate Addons wheel files for TensorFlow 1 and TensorFlow 2.

The wheel file has a name of the form:

ipu_tensorflow_addons-[ver]+X+X+X-X-X-X.whl

where [ver] is the version of the Graphcore port of TensorFlow 2. An example of the Addons wheel file for TensorFlow 2.6 for the IPU for Poplar SDK 3.0 is:

ipu_tensorflow_addons-2.6.3+gc3.0.0+236851+2e46901-py3-none-any.whl

Install the IPU TensorFlow 2 Addons wheel using the following command:

$ python -m pip install ${POPLAR_SDK_ENABLED?}/../ipu_tensorflow_addons-2.*.whl

POPLAR_SDK_ENABLED is the location of the Poplar SDK defined when the SDK was enabled. The ? ensures that an error message is displayed if Poplar has not been enabled.

3.4. Clone the Graphcore examples

You may need to clone the Graphcore examples repository on some systems as detailed in Table 3.2.

If you don’t need to clone the examples repository, then go straight to Section 3.5, Define environment variable.

Table 3.2 Systems that need the Graphcore tutorials and examples repositories to be cloned

System

Clone repos?

Comment

Pod system

Yes

You can clone the tutorials and examples repos in any location.

Graphcloud

Yes

You can clone the tutorials and examples repos in any location.

Gcore Cloud

No

The tutorials and examples have already been cloned in ~/graphcore/tutorials and ~/graphcore/examples respectively.

You can clone the examples repository into a location of your choice.

To clone the examples repository for the latest version of the Poplar SDK:

$ cd ~/[base_dir]
$ git clone https://github.com/graphcore/examples.git

where [base_dir] is a location of your choice. This will install the contents of the examples repository under ~/[base_dir]/examples. The tutorials are in ~/[base_dir]/examples/tutorials.

Note

If you are using a version of the Poplar SDK prior to version 3.2, then refer to Section A, Install examples and tutorials for older Poplar SDK versions for how to install examples and tutorials.

3.5. Define environment variable

In order to simplify running the tutorials, we define the environment variable POPLAR_TUTORIALS_DIR that points to the location of the cloned tutorials.

$ export POPLAR_TUTORIALS_DIR=~/[base_dir]/examples/tutorials

[base_dir] is the location where you installed the Graphcore tutorials.

3.6. Run the application

This section describes how to run a simple application, the MNIST example, using TensorFlow 2.

  1. Install example requirements

    You can now install the requirements that the model needs.

    $ cd $POPLAR_TUTORIALS_DIR/simple_applications/tensorflow2/mnist/
    $ pip install -r requirements.txt
    
  2. Run example

You run the code with the command:

$ python3 mnist.py

The example has no command line options.

If the code has run successfully, you should see an output similar to that in Listing 3.1.

Listing 3.1 Example of output for TensorFlow 2 application.
 2022-01-10 12:20:09.746730: I tensorflow/compiler/plugin/poplar/driver/poplar_platform.cc:44] Poplar version: 2.3.0 (d9e4130346) Poplar package: 88f485e763
 2022-01-10 12:20:11.195463: I tensorflow/compiler/jit/xla_cpu_device.cc:41] Not creating XLA devices, tf_xla_enable_xla_devices not set
 2022-01-10 12:20:11.435997: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:196] None of the MLIR optimization passes are enabled (registered 0 passes)
 2022-01-10 12:20:11.436536: I tensorflow/core/platform/profile_utils/cpu_utils.cc:112] CPU Frequency: 2245780000 Hz
 2022-01-10 12:20:12.922858: I tensorflow/compiler/plugin/poplar/driver/poplar_executor.cc:1714] Device /device:IPU:0 attached to IPU: 0
 2022-01-10 12:20:13.609918: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:116] None of the MLIR optimization passes are enabled (registered 2)
 Epoch 1/4
 Compiling module a_inference_train_function_513__XlaMustCompile_true_config_proto___n_007_n_0...02_001_000__executor_type____.380:
 [##################################################] 100% Compilation Finished [Elapsed: 00:00:15.4]
 2022-01-10 12:20:29.517778: I tensorflow/compiler/jit/xla_compilation_cache.cc:347] Compiled cluster using XLA!  This line is logged at most once for the lifetime of the process.
 2000/2000 [==============================] - 18s 9ms/step - loss: 0.9729
 Epoch 2/4
 2000/2000 [==============================] - 1s 533us/step - loss: 0.3478
 Epoch 3/4
 2000/2000 [==============================] - 1s 610us/step - loss: 0.2876
 Epoch 4/4
 2000/2000 [==============================] - 1s 595us/step - loss: 0.2545

You have run an application that demonstrates how to use the IPU to train a simple 2-layer, fully-connected model on the MNIST dataset using TensorFlow 2.

3.7. Exit the virtual environment

When you are done, exit the Python virtual environment.

$ deactivate

3.8. Try out other applications

The examples repo contains other tutorials and applications you can try. See Section 4, Next steps for more information.