Skip to content

Commit

Permalink
Add docs build workflow (#1319)
Browse files Browse the repository at this point in the history
* add workflow to test docs building.

* fix TrainingMethod path in docsrc.

---------

Co-authored-by: reuvenp <[email protected]>
  • Loading branch information
reuvenperetz and reuvenp authored Jan 8, 2025
1 parent e65d561 commit 3f04be7
Show file tree
Hide file tree
Showing 3 changed files with 39 additions and 4 deletions.
29 changes: 29 additions & 0 deletions .github/workflows/build_docs.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,29 @@
name: Build docs
on:
workflow_dispatch: # Allow manual triggers
schedule:
- cron: 0 0 * * *
pull_request:
branches:
- main

concurrency:
group: ${{ github.workflow }}-${{ github.ref }}
cancel-in-progress: ${{ github.ref != 'refs/heads/main' }}

jobs:
build:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
- name: Install Python 3
uses: actions/setup-python@v5
with:
python-version: 3.11
- name: Install dependencies
run: |
python -m pip install --upgrade pip
pip install -r requirements.txt
pip install sphinx==5.0.2 sphinx-autodoc-typehints==1.18.3 torch==2.5.* torchvision tensorflow==2.15.*
- name: Build docs
run: cd docsrc && make github SPHINXOPTS="-W --keep-going"
5 changes: 1 addition & 4 deletions docsrc/source/api/api_docs/modules/qat_config.rst
Original file line number Diff line number Diff line change
Expand Up @@ -10,10 +10,7 @@ qat_config Module

TrainingMethod
================================
**Select a QAT training method:**

.. autoclass:: model_compression_toolkit.qat.TrainingMethod

In order to select a training method, please visit the :ref:`trainable_infrastructure API.<ug-trainable_infrastructure>`

|
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -36,6 +36,15 @@ It adds to the base quantizer a get_config and from_config functions to enable l

.. autoclass:: model_compression_toolkit.trainable_infrastructure.BasePytorchTrainableQuantizer



TrainingMethod
================================
**Select a training method:**

.. autoclass:: model_compression_toolkit.trainable_infrastructure.TrainingMethod


TrainableQuantizerWeightsConfig
=================================
This configuration object contains the necessary attributes for configuring a weights trainable quantizer.
Expand Down

0 comments on commit 3f04be7

Please sign in to comment.