Navigate models

모델 주는 FuriosaAI NPU를 사용해서 가속할 수 있는 모델들을 쉽게 사용할 수 있도록 모아놓은 프로젝트입니다. 먼저, 사용 가능한 모델들을 살펴보겠습니다.

In [1]:

from furiosa.models import vision


print(dir(vision))

from furiosa.models import vision print(dir(vision))

['EfficientNetB0', 'EfficientNetV2s', 'ResNet50', 'SSDMobileNet', 'SSDResNet34', 'YOLOv5l', 'YOLOv5m']

In [2]:

# Or you can use the Command line tool
! furiosa-models list

# Or you can use the Command line tool ! furiosa-models list

+-----------------+------------------------------+----------------------+-------------------------+
|   Model name    |      Model description       |      Task type       | Available postprocesses |
+-----------------+------------------------------+----------------------+-------------------------+
|    ResNet50     |   MLCommons ResNet50 model   | Image Classification |         Python          |
|  SSDMobileNet   | MLCommons MobileNet v1 model |   Object Detection   |      Python, Rust       |
|   SSDResNet34   | MLCommons SSD ResNet34 model |   Object Detection   |      Python, Rust       |
|     YOLOv5l     |      YOLOv5 Large model      |   Object Detection   |          Rust           |
|     YOLOv5m     |     YOLOv5 Medium model      |   Object Detection   |          Rust           |
| EfficientNetB0  |    EfficientNet B0 model     | Image Classification |         Python          |
| EfficientNetV2s |    EfficientNetV2-s model    | Image Classification |         Python          |
+-----------------+------------------------------+----------------------+-------------------------+

그 중 하나인 ResNet50의 인스턴스를 만들고 Model 객체에 관해 자세히 알아보겠습니다.

In [3]:

model = vision.ResNet50()
print(model)

print("Static fields:", list(model.model_fields.keys()))
print("Lazy loaded fields:", list(model.model_computed_fields.keys()))

model = vision.ResNet50() print(model) print("Static fields:", list(model.model_fields.keys())) print("Lazy loaded fields:", list(model.model_computed_fields.keys()))

name='ResNet50' task_type=<ModelTaskType.IMAGE_CLASSIFICATION: 'IMAGE_CLASSIFICATION'> format=<Format.ONNX: 'ONNX'> family='ResNet' version='v1.5' metadata=Metadata(description='ResNet50 v1.5 int8 ImageNet-1K', publication=Publication(authors=None, title=None, publisher=None, date=None, url='https://arxiv.org/abs/1512.03385.pdf')) tags=None
Static fields: ['name', 'task_type', 'format', 'family', 'version', 'metadata', 'tags', 'preprocessor', 'postprocessor']
Lazy loaded fields: ['origin', 'tensor_name_to_range']

In [4]:

# You can see the static fields in Command line tool too
! furiosa-models desc ResNet50

# You can see the static fields in Command line tool too ! furiosa-models desc ResNet50

name: ResNet50
format: ONNX
family: ResNet
version: v1.5
metadata:
  description: ResNet50 v1.5 int8 ImageNet-1K
  publication:
    url: https://arxiv.org/abs/1512.03385.pdf
task type: Image Classification
available postprocess versions: Python

Actually there's one more hidden field (or, a method)

In [5]:

_ = model.model_source(num_pe=2)

_ = model.model_source(num_pe=2)