README_en.md

resnet50_vd_10w

Module Name	resnet50_vd_10w
Category	image classification
Network	ResNet_vd
Dataset	Baidu Dataset
Fine-tuning supported or not	No
Module Size	92MB
Latest update date	-
Data indicators	-

I.Basic Information

• Module Introduction

  • ResNet proposed a residual unit to solve the problem of training an extremely deep network, and improved the prediction accuracy of models. ResNet-vd is a variant of ResNet. This module is based on ResNet_vd, trained on Baidu dataset(consists of 100 thousand classes, 40 million pairs of data), and can predict an image of size 2242243.

II.Installation

• 1、Environmental Dependence

  • paddlepaddle >= 1.6.2

  • paddlehub >= 1.6.0 | How to install PaddleHub

• 2、Installation

  • $ hub install resnet50_vd_10w

  • In case of any problems during installation, please refer to: Windows_Quickstart | Linux_Quickstart | Mac_Quickstart

III.Module API Prediction

• 1、Prediction Code Example

  • import paddlehub as hub
    import cv2
    
    classifier = hub.Module(name="resnet50_vd_10w")
    input_dict, output_dict, program = classifier.context(trainable=True)

• 2、API

  • def context(trainable=True, pretrained=True)

    • Parameters

      • trainable (bool): whether parameters are trainable；
        
      • pretrained (bool): whether load the pre-trained model.

    • Return

      • inputs (dict): model inputs，key is 'image', value is the image tensor；
        
      • outputs (dict): model outputs，key is 'classification' and 'feature_map'，values：
        • classification (paddle.fluid.framework.Variable): classification result；
        • feature_map (paddle.fluid.framework.Variable): feature map extracted by model.
      • context_prog(fluid.Program): computation graph, used for transfer learning.

  • def save_inference_model(dirname,
                             model_filename=None,
                             params_filename=None,
                             combined=True)

    • Parameters
      • dirname: output dir for saving model;
        
      • model_filename: filename of model, default is __model__;
        
      • params_filename: filename of parameters，default is __params__(only effective when combined is True);
        
      • combined: whether save parameters into one file

V.Release Note

• 1.0.0

  First release

  • $ hub install resnet50_vd_10w==1.0.0