Upload Model

upload_model(self, model_name: str, task: netspresso.compressor.client.utils.enum.Task, framework: netspresso.compressor.client.utils.enum.Framework, file_path: str, input_shapes: List[Dict[str, int]] = []) → netspresso.compressor.core.model.Model

Upload a model for compression.

Parameters

model_name (str) – The name of the model.
task (Task) – The task of the model.
framework (Framework) – The framework of the model.
file_path (str) – The file path where the model is located.
input_shapes (List[Dict[str, int]], optional) – Input shapes of the model. Defaults to [].

Raises

e – If an error occurs while uploading the model.

Returns

Uploaded model object.

Return type

Model

Details of Parameters

Task

class Task[source]: An enumeration.

Available Task

Name	Description
IMAGE_CLASSIFICATION	Image Classification
OBJECT_DETECTION	Object Detection
IMAGE_SEGMENTATION	Image Segmentation
SEMANTIC_SEGMENTATION	Semantic Segmentation
INSTANCE_SEGMENTATION	Instance Segmentation
PANOPTIC_SEGMENTATION	Panoptic Segmentation
OTHER	Other

Example

from netspresso.compressor import Task

TASK = Task.IMAGE_CLASSIFICATION

Framework

class Framework[source]: An enumeration.

Available Framework

Name	Description
TENSORFLOW_KERAS	TensorFlow-Keras
PYTORCH	PyTorch GraphModule
ONNX	ONNX

Example

from netspresso.compressor import Framework

FRAMEWORK = Framework.TENSORFLOW_KERAS

Note

ONNX (.onnx)
- Supported version: Pytorch >= 1.11.x, ONNX >= 1.10.x.
- If a model is defined in Pytorch, it should be converted into the ONNX format before being uploaded.
- How-to-guide for the conversion of PyTorch into ONNX.
PyTorch GraphModule (.pt)
- Supported version: Pytorch >= 1.11.x.
- If a model is defined in Pytorch, it should be converted into the GraphModule before being uploaded.
- The model must contain not only the status dictionary but also the structure of the model (do not use state_dict).
- How-to-guide for the conversion of PyTorch into GraphModule.
TensorFlow-Keras (.h5, .zip)
- Supported version: TensorFlow 2.3.x ~ 2.8.x.
- Custom layer must not be included in Keras H5 (.h5) format.
- The model must contain not only weights but also the structure of the model (do not use save_weights).
- If there is a custom layer in the model, please upload TensorFlow SavedModel format (.zip).

Input Shapes

Note

For input shapes, use the same values that you used to train the model.
- If the input shapes of the model is dynamic, input shapes is required.
- If the input shapes of the model is static, input shapes is not required.

For example, batch=1, channel=3, height=768, width=1024.

input_shapes = [{"batch": 1, "channel": 3, "dimension": [768, 1024]}]

Currently, only single input models are supported.

Details of Returns

class Model(model_id: str, model_name: str, task: str, framework: str, model_size: float, flops: float, trainable_parameters: float, non_trainable_parameters: float, number_of_layers: int, input_shapes: List[netspresso.compressor.core.model.InputShape] = <factory>)[source]

Represents a uploaded model.

model_id

The ID of the model.

Type: str

model_name

The name of the model.

Type: str

task

The task of the model.

Type: Task

framework

The framework of the model.

Type: Framework

input_shapes

The input shapes of the model.

InputShape Attributes:

batch (int): The batch size of the input tensor.
channel (int): The number of channels in the input tensor.
dimension (List[int]): The dimensions of the input tensor.

Type: List[InputShape]

model_size

The size of the model.

Type: float

flops

The FLOPs (floating point operations) of the model.

Type: float

trainable_parameters

The number of trainable parameters in the model.

Type: float

non_trainable_parameters

The number of non-trainable parameters in the model.

Type: float

number_of_layers

The number of layers in the model.

Type: float

Example

from netspresso.compressor import ModelCompressor, Task, Framework


compressor = ModelCompressor(email="YOUR_EMAIL", password="YOUR_PASSWORD")
model = compressor.upload_model(
    model_name="YOUR_MODEL_NAME",
    task=Task.IMAGE_CLASSIFICATION,
    framework=Framework.TENSORFLOW_KERAS,
    file_path="YOUR_MODEL_PATH", # ex) ./model.h5
    input_shapes="YOUR_MODEL_INPUT_SHAPES",  # ex) [{"batch": 1, "channel": 3, "dimension": [32, 32]}]
)

Output

>>> model
Model(
    model_id="5eeb0edb-57d2-4a20-adf4-a6c05516015d",
    model_name="YOUR_MODEL_NAME",
    task="image_classification",
    framework="tensorflow_keras",
    input_shapes=[InputShape(batch=1, channel=3, dimension=[32, 32])],
    model_size=12.9641,
    flops=92.8979,
    trainable_parameters=3.3095,
    non_trainable_parameters=0.0219,
    number_of_layers=0,
)