Update README for installing the published package

This commit is contained in:
Douglas Orr 2017-10-06 14:46:38 +01:00
Родитель 87ff98dd7f
Коммит abe71b8a5e
2 изменённых файлов: 21 добавлений и 16 удалений

Просмотреть файл

@ -14,11 +14,13 @@ It is really quite hard to test language model performance. Some models output p
## Getting Started
We include support for building a Docker image that makes it easier to get started (which also documents [our dependencies](Dockerfile), in case you want to get set up natively).
Install LMChallenge from the published Python package:
```bash
docker build --rm -t lmchallenge .
```
pip3 install --user lmchallenge
Or from this repository:
python3 setup.py install --user
## The Challenges

Просмотреть файл

@ -14,24 +14,17 @@ This project has adopted the [Microsoft Open Source Code of Conduct](https://ope
For more information see the [Code of Conduct FAQ](https://opensource.microsoft.com/codeofconduct/faq/) or
contact [opencode@microsoft.com](mailto:opencode@microsoft.com) with any additional questions or comments.
## Tests
## Developing & testing
These are currently very much incomplete (please help fix this), but they are run with:
We include support for building a Docker image that makes it easier to get started (which also documents [our dependencies](Dockerfile), in case you want to get set up natively).
Tests are currently very much incomplete (please help fix this), but they are run with:
./scripts/run build
./scripts/run -i prod test
./scripts/run check
## Publishing
1. ensure you've run the pre-publish checks `./scripts/run check`
2. check that you're happy with `version.txt`
3. `python3 setup.py sdist upload -r pypi`
- you must first set up pypi & GPG credentials
4. `git push origin HEAD:refs/tags/$(cat version.txt)`
5. update, commit & push `version.txt`
## Documentation
### Documentation
You can build the documentation with the development Docker image.
@ -50,6 +43,16 @@ For now, we don't use an internal registry, but you can send things around manua
scp lmchallenge-doc.tar.gz ...
gzip -cd lmchallenge-doc.tar.gz | docker load
### Publishing
1. ensure you've run the pre-publish checks `./scripts/run check`
2. check that you're happy with `version.txt`
3. `python3 setup.py sdist upload -r pypi`
- you must first set up pypi & GPG credentials
4. `git push origin HEAD:refs/tags/$(cat version.txt)`
5. update, commit & push `version.txt`
## Ideas
### `rank` command