Skip to content

Latest commit

 

History

History
83 lines (54 loc) · 2.25 KB

mobilenet-v2-pytorch.md

File metadata and controls

83 lines (54 loc) · 2.25 KB
Raw

mobilenet-v2-pytorch

Use Case and High-Level Description

MobileNet V2 is image classification model pretrained on ImageNet dataset. This is a PyTorch implementation of MobileNetV2 architecture as described in the paper "Inverted Residuals and Linear Bottlenecks: Mobile Networks for Classification, Detection and Segmentation".

The model input is a blob that consists of a single image of "1x3x224x224" in RGB order.

The model output is typical object classifier for the 1000 different classifications matching with those in the ImageNet database.

Example

See here

Specification

Metric Value
Type Classification
GFLOPs 0.615
MParams 3.489
Source framework PyTorch*

Accuracy

Metric Original model Converted model
Top 1 71.8% 71.8%
Top 5 90.396% 90.396%

Performance

Input

Original model

Image, name - data, shape - 1,3,224,224, format is B,C,H,W where:

  • B - batch size
  • C - channel
  • H - height
  • W - width

Channel order is RGB. Mean values - [123.675,116.28,103.53], scale value - [58.624,57.12,57.375]

Converted model

Image, name - data, shape - 1,3,224,224, format is B,C,H,W where:

  • B - batch size
  • C - channel
  • H - height
  • W - width

Channel order is BGR

Output

Original model

Object classifier according to ImageNet classes, name - prob, shape - 1,1000, output data format is B,C where:

  • B - batch size
  • C - Predicted probabilities for each class in [0, 1] range

Converted model

Object classifier according to ImageNet classes, name - prob, shape - 1,1000, output data format is B,C where:

  • B - batch size
  • C - Predicted probabilities for each class in [0, 1] range

Legal Information

The original model is distributed under the Apache License, Version 2.0. A copy of the license is provided in APACHE-2.0.txt.