Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Incorrect docs for --model-store option

See original GitHub issue

Docs for the torchserve command line at https://github.com/pytorch/serve/blob/master/docs/server.md make the claim:

model-store: optional, A location where models are stored by default, all models in this location are loaded, the model name is same as archive or folder name.

This is in reference to the --model-store argument. This is incorrect; if it were true, I should be able to call torchserve --start --model-store model_store (where I have multiple models in the folder model_store) and get endpoints - but I get only 404s when trying to call them.

Issue Analytics

State:
Created 4 years ago
Comments:6 (4 by maintainers)

Top GitHub Comments

1reaction

mycpuorgcommented, Feb 19, 2020

@fbbradheintz We are working on checkpoint/restart APIs. As of current implementation, these models have to be registered explicitly, each. Impl of Checkpoint/Restart will give us this functionality in effect.

0reactions

fbbradheintzcommented, Feb 28, 2020

This looks great, and works as documented. Please merge when ready.

Top Results From Across the Web

modelstore

modelstore is a Python library that enables you to create a registry for your machine learning models. This enables you to automatically:.

5. Advanced configuration - PyTorch

However, if you want to customize TorchServe, the configuration options ... –model-store Overrides the model_store property in config.properties file.

galaxy.model.store — Galaxy Project 20.09 documentation

Source code for galaxy.model.store ... [docs] def __init__(self, allow_edit=False, allow_library_creation=False, allow_dataset_object_edit=None): ...

DLR image classification model store - AWS IoT Greengrass

Requirements · Run the following command to open the Raspberry Pi configuration tool. sudo raspi-config · Select Interface Options. · Select Legacy camera...

Preparing Models - BentoML

Saved models are stored in BentoML's model store, which is a local file ... model signatures for runner inference "classify": { "batchable": False,...