MindSpore Hub Documents

MindSpore Hub is a pre-trained model application tool of the MindSpore ecosystem. It provides the following functions:

  • Plug-and-play model loading

  • Easy-to-use transfer learning

import mindspore
import mindspore_hub as mshub
from mindspore import set_context, GRAPH_MODE

set_context(mode=GRAPH_MODE,
                    device_target="Ascend",
                    device_id=0)

model = "mindspore/1.6/googlenet_cifar10"

# Initialize the number of classes based on the pre-trained model.
network = mshub.load(model, num_classes=10)
network.set_train(False)

# ...

Typical Application Scenarios

  1. Inference Validation

    With only one line of code, use mindspore_hub.load to load the pre-trained model.

  2. Transfer Learning

    After loading models using mindspore_hub.load, add an extra argument to load the feature extraction of the neural network. This makes it easier to add new layers for transfer learning.

  3. Model Releasing

    Release the trained model to MindSpore Hub according to the specified procedure for download and use.

API References