||3 days ago|
|argostrain||3 months ago|
|bin||1 week ago|
|docs||7 months ago|
|scripts||4 months ago|
|.gitignore||3 months ago|
|Dockerfile||7 months ago|
|LICENSE||2 years ago|
|MODEL_README.md||7 months ago|
|README.md||3 days ago|
|config.yml||7 months ago|
|data-index.json||3 days ago|
|requirements.txt||2 months ago|
|sample_data.py||1 year ago|
|setup.py||7 months ago|
Argos Train trains an OpenNMT PyTorch Transformer model and a SentencePiece tokenizer and packages them with Stanza data as an Argos Translate package. Argos Translate packages, which are zip archives with a .argosmodel extension, can be used with Argos Translate, LibreTranslate, and Dot Lexicon.
Pre-trained Argos Translate packages are available for download. If you have trained packages you're willing to share please get in contact so that they can be published on the Argos Translate package index.
$ su argosopentech $ source ~/argos-train-init ... $ argos-train From code (ISO 639): en To code (ISO 639): es From name: English To name: Spanish Version: 1.0 ... Package saved to /home/argosopentech/argos-train/run/en_es.argosmodel
To train a model with custom data add your data to
data-index.json after running
argos-train-init with a link to download your custom data package. Data packages are zipped directories with a .argosdata extension that contain a
target file with parallel data in corresponding lines and a
Docker image available at argosopentech/argostrain.
docker run -it argosopentech/argostrain /bin/bash
CUDA required, tested on vast.ai.
Licensed under either the MIT or CC0 License (same as Argos Translate).