Skip to content

Latest commit

 

History

History
75 lines (49 loc) · 2.08 KB

efficientnet-b0.md

File metadata and controls

75 lines (49 loc) · 2.08 KB
Raw

efficientnet-b0

Use Case and High-Level Description

The efficientnet-b0 model is one of the EfficientNet models designed to perform image classification. This model was pretrained in TensorFlow*. All the EfficientNet models have been pretrained on the ImageNet* image database. For details about this family of models, check out the TensorFlow Cloud TPU repository.

Example

Specification

Metric Value
Type Classification
GFLOPs 0.819
MParams 5.268
Source framework TensorFlow*

Accuracy

Metric Original model Converted model
Top 1 75.70% 75.70%
Top 5 92.76% 92.76%

Performance

Input

Original Model

Image, name - image, shape - [1x224x224x3], format is [BxHxWxC], where:

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

Channel order is RGB.

Converted Model

Image, name - sub/placeholder_port_0, shape - [1x3x224x224], format is [BxCxHxW], where:

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

Channel order is BGR.

Output

Original Model

Object classifier according to ImageNet classes, name - logits, 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 - efficientnet-b0/model/head/dense/MatMul, 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-TF-TPU.txt.