# VGGNet

**Repository Path**: cuzjx/VGGNet

## Basic Information

- **Project Name**: VGGNet
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2019-12-20
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

This is a Tensorflow implementation of VGG Net.
The architecture follows the 16-layer architecture (vgg16) from the original paper:
https://arxiv.org/pdf/1409.1556.pdf

The default is to load pre-trained weights. To change this option, change the value of the flag named "use_pre_trained" under VGGNet_train.py.
If not using the pre-trained weights, initial learning rate should be increased (usually starting from 0.01).

The classification layer follows the labeling order that is defined in the file ImageNet_classes.txt. This is not the same ordering as in the original ILSVRC2012 devkit.