# Modelhub
A CLI tool to manage your models.
## Commands
It takes a TensorFlow SaveModel file, uploads it with version and meta info, and checkouts it to local.
- info model_name[@version_number]
- create model_name -m “comment”
- submit model_name -m “comment” [–batching_config=batching_config.txt] model_path
- checkout model_name[@version_number]
- deploy model_name[@version_number] [CPU/GPU] [online/offline] [instance-number] [offline]
## NTF models
ModelHub ogirinally supports uploading only TensorFlow-based models. From 0.0.8, it supports NTF models.
To prepare NTF models, you need create a new folder containing two subfolders: variables and assets
varibles: storing trained models
assets: storing any other data files, for example, feature dictionaries.
## Python API pass get model path by model name [and version]
checkout model
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
File details
Details for the file modelhub-3.0.10.tar.gz.
File metadata
- Download URL: modelhub-3.0.10.tar.gz
- Upload date:
- Size: 74.2 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/1.11.0 pkginfo/1.4.2 requests/2.19.1 setuptools/38.0.0 requests-toolbelt/0.8.0 tqdm/4.23.4 CPython/3.6.5
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
9127a01833600764982da3812d345fbfd4d7980be782b9e055d9e98d78cbf336
|
|
| MD5 |
0e870c6585dfc0538e30a40a069ff4eb
|
|
| BLAKE2b-256 |
9966919d69ab3d964b63782b0d9df124186b739052379c9e190e3355931aebe1
|
