-
大小: 961KB文件类型: .rar金币: 2下载: 0 次发布日期: 2021-05-23
- 语言: 其他
- 标签: tensorflow
资源简介
下载深度学习的VGG19网络参数,有下载地址,VGG19网络参数参数下载放在指定的位置,即可运行

代码片段和文件信息
import tensorflow as tf
import numpy as np
import utils
import vgg19
import style_transfer
import os
import argparse
“““parsing and configuration“““
def parse_args():
desc = “Tensorflow implementation of ‘Image style Transfer Using Convolutional Neural Networks“
parser = argparse.ArgumentParser(description=desc)
parser.add_argument(‘--model_path‘ type=str default=‘pre_trained_model‘ help=‘The directory where the pre-trained model was saved‘)
parser.add_argument(‘--content‘ type=str default=‘images/a12345.jpg‘ help=‘File path of content image (notation in the paper : p)‘)
parser.add_argument(‘--style‘ type=str default=‘images/kandinsky.jpg‘ help=‘File path of style image (notation in the paper : a)‘)
parser.add_argument(‘--output‘ type=str default=‘result_a12345_kandinsky.jpg‘ help=‘File path of output image‘)
parser.add_argument(‘--loss_ratio‘ type=float default=1e-3 help=‘Weight of content-loss relative to style-loss‘)
parser.add_argument(‘--content_layers‘ nargs=‘+‘ type=str default=[‘conv4_2‘] help=‘VGG19 layers used for content loss‘)
parser.add_argument(‘--style_layers‘ nargs=‘+‘ type=str default=[‘relu1_1‘ ‘relu2_1‘ ‘relu3_1‘ ‘relu4_1‘ ‘relu5_1‘]
help=‘VGG19 layers used for style loss‘)
parser.add_argument(‘--content_layer_weights‘ nargs=‘+‘ type=float default=[1.0] help=‘Content loss for each content is multiplied by corresponding weight‘)
parser.add_argument(‘--style_layer_weights‘ nargs=‘+‘ type=float default=[.2.2.2.2.2]
help=‘style loss for each content is multiplied by corresponding weight‘)
parser.add_argument(‘--initial_type‘ type=str default=‘content‘ choices=[‘random‘‘content‘‘style‘] help=‘The initial image for optimization (notation in the paper : x)‘)
parser.add_argument(‘--max_size‘ type=int default=512 help=‘The maximum width or height of input images‘)
parser.add_argument(‘--content_loss_norm_type‘ type=int default=3 choices=[123] help=‘Different types of normalization for content loss‘)
parser.add_argument(‘--num_iter‘ type=int default=1000 help=‘The number of iterations to run‘)
return check_args(parser.parse_args())
“““checking arguments“““
def check_args(args):
try:
assert len(args.content_layers) == len(args.content_layer_weights)
except:
print (‘content layer info and weight info must be matched‘)
return None
try:
assert len(args.style_layers) == len(args.style_layer_weights)
except:
print(‘style layer info and weight info must be matched‘)
return None
try:
assert args.max_size > 100
except:
print (‘Too small size‘)
return None
try:
assert os.path.exists(args.content)
except:
print(‘There is no %s‘%args.content)
return None
try:
assert os.path.exists(args.style)
except:
print(‘There is no %s
属性 大小 日期 时间 名称
----------- --------- ---------- ----- ----
文件 185 2018-09-27 12:47 tensorflow-st
文件 324 2018-09-27 11:32 tensorflow-st
文件 459 2018-09-27 11:36 tensorflow-st
文件 11111 2018-09-27 14:00 tensorflow-st
文件 202426 2017-02-17 09:12 tensorflow-st
文件 101993 2017-02-17 09:12 tensorflow-st
文件 144730 2017-02-17 09:12 tensorflow-st
文件 205803 2017-02-17 09:12 tensorflow-st
文件 216723 2017-02-17 09:12 tensorflow-st
文件 101689 2017-02-17 09:12 tensorflow-st
文件 5955 2017-02-17 09:12 tensorflow-st
文件 5922 2018-09-27 14:00 tensorflow-st
文件 5787 2017-02-17 09:12 tensorflow-st
文件 2816 2017-02-17 09:12 tensorflow-st
文件 2567 2018-09-27 11:54 tensorflow-st
文件 75 2018-09-27 14:15 tensorflow-st
文件 3758 2018-09-27 12:51 tensorflow-st
文件 1604 2018-09-27 12:51 tensorflow-st
文件 2964 2018-09-27 12:51 tensorflow-st
目录 0 2018-09-27 14:00 tensorflow-st
目录 0 2018-09-27 13:54 tensorflow-st
目录 0 2018-09-27 12:49 tensorflow-st
目录 0 2018-09-27 14:55 tensorflow-st
目录 0 2018-09-27 13:47 tensorflow-st
目录 0 2018-09-27 14:00 tensorflow-st
----------- --------- ---------- ----- ----
1016891 25
- 上一篇:wiki.zh.text.model
- 下一篇:基于keras实现的LSTM网络
相关资源
- TensorFlow1.4 官方手册
- MSCNN_dehaze.rar
- 基于双向LSTM+tensorflow中文分词
- TensorFlow Machine Learning Cookbook+无码高清
- Hands-On Machine Learning with Scikit-Learn an
- TensorFlow 官方文档中文版高清无码PD
- 基于tensorflow的深度学习图像分类案例
- 基于Tensorflow多层神经网络的MNIST手写
- TensorFlow Tutorial
- TensorFlow_2.0快速应用指南英文.tar
- tensorflow图像风格迁移代码
- TensorFlow Machine Learning Cookbook
- Learning TensorFlow: A Guide to Building Deep
- tensorflow271134
- TensorFlow for Machine Intelligence 书籍源码
- 《Hands-On Machine Learning with Scikit-Learn
- Building Machine Learning Projects with Tensor
- TensorFlow Powerful Predictive Analytics with
- 北大tensorflow学习笔记
- LSTMandRNN.zip
- 使用TensorFlow搭建智能开发系统,自动
- Packt.TensorFlow.Machine.Learning.Cookbook.201
- tensorflow深度学习CNN智能识别车牌
- Tensorflow GNN实战.zip
- tensorflow从入门到精通
- 深度卷积生成对抗网络TensorFlow代码实
- TensorFlow_中文手册
- siamese mnist 孪生网络
- TensorFlow 中文文档 1.8 掘金 2018.5.19
- Tensorflow(一)训练自己的数据集——
评论
共有 条评论