Metadata-Version: 2.1
Name: hearvalidator
Version: 2021.0.6
Summary: Holistic Evaluation of Audio Representations (HEAR) 2021 -- Submission Validator
Home-page: https://github.com/hearbenchmark/hear-validator
Author: HEAR 2021 NeurIPS Competition Committee
Author-email: deep@neuralaudio.ai
License: Apache-2.0
Project-URL: Bug Tracker, https://github.com/hearbenchmark/hear-validator/issues
Project-URL: Source Code, https://github.com/hearbenchmark/hear-validator
Requires-Python: >=3.6, <3.9
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: tensorflow (>=2.0)
Requires-Dist: torch (>=1.7)
Provides-Extra: dev
Requires-Dist: pre-commit ; extra == 'dev'
Requires-Dist: black ; extra == 'dev'

# HEAR Benchmark Submission Validator

This package provides a command-line tool to verify that a python module follows the
HEAR [common API](https://hearbenchmark.com/hear-api.html).

For full details on the HEAR benchmark please visit https://hearbenchmark.com

### Installation

Tested with Python 3.7 and 3.8. Python 3.9 is not officially supported
because pip3 installs are very finicky, but it might work.

```python
pip install hearvalidator
```
This will install a command-line tool: `hear-validator`

### Usage

Let's validate the [HEAR naive baseline model](https://github.com/hearbenchmark/hear-baseline):
```
pip install hearbaseline
wget https://github.com/hearbenchmark/hear-baseline/raw/main/saved_models/naive_baseline.pt
hear-validator hearbaseline --model ./naive_baseline.pt
```

Optional arguments for `hear-validator`:

 * `--model`: Load the model weights from this path.
 * `--device`: Device to run validation on. If this isn't specified then this
 will run on a GPU if one is present. If no GPU is available then this will
 default to the CPU.
