# Example Name/Acronym (Full Expanded Name if Applicable)
Give a one-line summary of the example and include images, if relevant.

(Table that summarises app usage and capability)
| Framework | Domain | Model | Datasets | Tasks | Training | Inference | Reference |
|-----------|--------|-------|----------|-------|----------|-----------|-----------|
| one of them | NLP | name | list| Q&A / Classification, etc.. | <p style="text-align: center;">✅ <br> Min. 1 IPU (POD4) required  | <p style="text-align: center;">✅ <br> Min. 1 IPU (POD4) required | POD4/POD16/POD64 | link to paper/original implementation|

## (Optional) Extended introduction
Give a short description (~3-10 lines) about the example, the methods or technology used, applications, tasks etc. This should fill the role of an extended introduction to the example.

## Overview
Before you can run this model for training or inference you will need to:

1. Enable the Poplar SDK (see [Poplar SDK setup](#poplar-sdk-setup))
2. Install the system and Python requirements (see [Environment setup](#environment-setup))
3. Download the XX dataset (See [Dataset setup](#dataset-setup))
4. etc.

### Poplar SDK setup
To check if your Poplar SDK has already been enabled, run:
```bash
 echo $POPLAR_SDK_ENABLED
 ```

If no path is printed, then follow these steps to enable the Poplar SDK:
1. Navigate to your Poplar SDK root directory

2. Enable the Poplar SDK with:
```bash
cd poplar-<OS version>-<SDK version>-<hash>
. enable
```

Detailed instructions on enabling the Poplar SDK are available in the [<framework> quick start guide] (link from below -- Note: All quick starts have instructions to enable the SDK, so only refer to the quick start for the framework you are using. Do not link to the Poplar Quick Start.)
PyTorch - https://docs.graphcore.ai/projects/pytorch-quick-start
TensorFlow 2 - https://docs.graphcore.ai/projects/tensorflow2-quick-start
PopXL - https://docs.graphcore.ai/projects/popxl/en/latest/setup_quick_start.html.


## Environment setup
To prepare your environment, follow these steps:

1. Create and activate a Python3 virtual environment:
```bash
python3 -m venv <venv name>
source <venv path>/bin/activate
```

2. Install the framework-specific wheels from the SDK directory and validate the installation:

[If you are using PopTorch:]
```bash
pip3 install ${POPLAR_SDK_ENABLED?}/../poptorch-*.whl
python3 -c "import poptorch; print(poptorch.__version__)"
```

[If you are using TensorFlow 2:]
```bash
pip3 install ${POPLAR_SDK_ENABLED?}/../tensorflow-2.*+amd_*.whl
pip3 install --force-reinstall --no-deps ${POPLAR_SDK_ENABLED?}/../keras-2.*.whl
pip3 install ${POPLAR_SDK_ENABLED?}/../ipu_tensorflow_addons-2.*.whl
python3 -c "from tensorflow.python import ipu"
python3 -c "import keras"
python3 -c "from ipu_tensorflow_addons.keras import layers"
```

Detailed instructions on setting up your <framework> environment are available in the [<framework> quick start guide](<link from below>)
PyTorch - https://docs.graphcore.ai/projects/pytorch-quick-start
TensorFlow 2 - https://docs.graphcore.ai/projects/tensorflow2-quick-start
PopXL - https://docs.graphcore.ai/projects/popxl/en/latest/setup_quick_start.html.


4. Navigate to this example's root directory

5. Pre-requirements build steps
```bash
make
```

5. Install the Python requirements with:
```bash
pip3 install -r requirements.txt
```

6. Additional build steps:
```bash
make
```

### Additional build steps
Describe what these steps are, and what their purpose is.
```bash
sudo apt-get install xxx
make clean && make
```

## Dataset setup

### Dataset name
Describe (1-2 lines) what this dataset is and what it is used for.

Disk space required: XXGB
```bash
bash prep_data.sh etc.
```
Include the file structure tree for dataset in tree form, using "tree -L 3 --filelimit 5"
dataset_dir
├── README.md
├── train
|   ├── subdir_1
|   └── subdir_2
├── validate
    ├── subdir_1
    └── subdir_2

### Another dataset name
Describe (1-2 lines) what this dataset is and what it is used for.

Disk space required: XXGB
```bash
bash prep_data.sh etc.
```
Include the file structure tree for dataset in tree form, using "tree -L 3 --filelimit 5"
dataset_dir
├── README.md
├── train
|   ├── subdir_1
|   └── subdir_2
├── validate
    ├── subdir_1
    └── subdir_2


## Running and benchmarking

To run a tested and optimised configuration and to reproduce the performance shown on our [performance results page](https://www.graphcore.ai/performance-results), use the `examples_utils` module (installed automatically as part of the environment setup) to run one or more benchmarks. The benchmarks are provided in the `benchmarks.yml` file in this example's root directory.

For example:

```bash
python3 -m examples_utils benchmark --spec <path to benchmarks.yml file>
```

Or to run a specific benchmark in the `benchmarks.yml` file provided:

```bash
python3 -m examples_utils benchmark --spec <path to benchmarks.yml file> --benchmark <name of benchmark>
```

For more information on using the `examples-utils` benchmarking module, please refer to the [module README](https://github.com/graphcore/examples-utils/blob/master/examples_utils/benchmarks/README.md).


## Custom training
Give basic info on how to run your own custom training with commands

## Custom Inference
Give basic info on how to run your own custom inference with commands

## Other features
Describe or demonstrate any other aspects of the app that you want to highlight that don't fit into the sections above.

## License

This application is licensed under [name of license].0. Please see the [LICENSE]<link to license> file for full details of the license conditions.

Describe if any content in the application doesn't fall under the above license, for example, images or extracts from research papers, and list details of the licenses of these.