Automatic Compression

Description

netspresso.compressor.v2.compressor.CompressorV2.automatic_compression(input_model_path, output_dir, input_shapes, framework=Framework.PYTORCH, compression_ratio=0.5)

Compress a model automatically based on the given compression ratio.

Parameters:

Name	Type	Description	Default
input_model_path	str	The file path where the model is located.	required
output_dir	str	The local path to save the compressed model.	required
input_shapes	List[Dict[str, int]]	Input shapes of the model.	required
framework	Framework	The framework of the model.	PYTORCH
compression_ratio	float	The compression ratio for automatic compression. Defaults to 0.5.	0.5

Raises:

Type	Description
e	If an error occurs while performing automatic compression.

Returns:

Name	Type	Description
CompressorMetadata	CompressorMetadata	Compress metadata.

Details of Parameters

Compression Ratio

Note

• As the compression ratio increases, you can get more lighter and faster compressed models, but with greater lost of accuracy.
• Therefore, it is necessary to find an appropriate ratio for your requirements. It might require a few trials and errors.
• The range of available values is as follows.

\[0 < \text{ratio} < 1\]

Example

from netspresso import NetsPresso


# Login with API key (recommended)
# Get your API token from: https://account.netspresso.ai/api-token
netspresso = NetsPresso(api_key="YOUR_API_KEY")

# Note: Email/password login will be deprecated soon
# netspresso = NetsPresso(email="YOUR_EMAIL", password="YOUR_PASSWORD")

compressor = netspresso.compressor_v2()
compressed_model = compressor.automatic_compression(
    input_shapes=[{"batch": 1, "channel": 3, "dimension": [224, 224]}],
    input_model_path="./examples/sample_models/graphmodule.pt",
    output_dir="./outputs/compressed/pytorch_automatic_compression_1",
    compression_ratio=0.5,
)