# 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
Release files for modelhub 3.0.10
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| modelhub-3.0.10.tar.gz | 74.2 kB | Details |
Release files / modelhub-3.0.10.tar.gz
| Download URL | modelhub-3.0.10.tar.gz |
|---|---|
| Size | 74.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
9127a01833600764982da3812d345fbfd4d7980be782b9e055d9e98d78cbf336
|
|
BLAKE2b-256 checksum How to use checksums |
9966919d69ab3d964b63782b0d9df124186b739052379c9e190e3355931aebe1
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is 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
|
