Metadata-Version: 2.1
Name: bentoutils
Version: 0.0.41
Summary: Utilities for working with BentoML
Home-page: https://github.com/markmo/bentoutils
Author: Mark Moloney
Author-email: m4rkmo@gmail.com
License: UNKNOWN
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
Requires-Dist: BentoML (==0.10.1)
Requires-Dist: boto3 (==1.16.43)
Requires-Dist: click (==7.1.2)
Requires-Dist: kubernetes (==12.0.1)
Requires-Dist: PyYAML (==5.3.1)
Requires-Dist: stringcase (==1.2.0)
Requires-Dist: text-unidecode (==1.3)

# bentoutils

## Contents

A) BentoML Framework (Custom Artifact) for [simpletransformers](https://github.com/ThilinaRajapakse/simpletransformers).

B) Console scripts for:

1. bentopack - package an existing pretrained model and save to the Model Registry

```
Usage: bentopack [OPTIONS]

Options:
  --module TEXT  fully qualified module name containing service to package
  --clz TEXT     class name of service to package
  --name TEXT    model name
  --path TEXT    directory path of pretrained model
  --help         Show this message and exit.
```

Example:
```
bentopack \
    --module TopicBentoService \        # python module containing service class
    --clz TopicBentoService \           # service class
    --name tm_train3_roberta_l_weigh \  # pretrained model name
    --path /srv/models/multilabel-topic # local path to pretrained model (excluding name)
```

2. deploy_to_knative - WIP


