使⽤tensorlayer时，出现了⼤量相关的参数设置，通⽤的参数设置如下:
task = 'dcgan'
flags = tf.app.flags
flags.DEFINE_string('task','dcgan','this task name')
flags.DEFINE_integer("epoch", 200, "Epoch to train [100]")
flags.DEFINE_float("learning_rate", 0.0002, "Learning rate of for adam [0.0002]")
flags.DEFINE_float("beta1", 0.5, "Momentum term of adam [0.5]")
flags.DEFINE_float("weight_decay", 1e-5, "Weight decay for l2 loss")
flags.DEFINE_float("pool_size", 50, 'size of image buffer that stores previously generated images, default: 50')
flags.DEFINE_integer("train_size", 3000, "The size of train images [np.inf]")
flags.DEFINE_integer("batch_size", 1, "The number of batch images [1] if we use InstanceNormLayer !")
flags.DEFINE_integer("image_size", 256, "The size of image to use (will be center cropped) [256]")
flags.DEFINE_integer("gf_dim", 32, "Size of generator filters in first layer")
flags.DEFINE_integer("df_dim", 64, "Size of discriminator filters in first layer")
# flags.DEFINE_integer("class_embedding_size", 5, "Size of class embedding")
flags.DEFINE_integer("output_size", 256, "The size of the output images to produce [64]")
flags.DEFINE_integer("sample_size", 64, "The number of sample images [64]")
flags.DEFINE_integer("c_dim", 3, "Dimension of image color. [3]")
flags.DEFINE_integer("sample_step", 500, "The interval of generating sample. [500]")
flags.DEFINE_integer("save_step", 200, "The interval of saveing checkpoints. [200]")
flags.DEFINE_string("dataset_dir", "spring2snow", "The name of dataset [horse2zebra, apple2orange, sunflower2daisy and etc]")
flags.DEFINE_string("checkpoint_dir", "/home/liuwenjie/deep_save/{}/ckpt".format(task), "Directory name to save the checkpoints [checkpoint]") flags.DEFINE_string("sample_dir", '/home/liuwenjie/deep_save/{}/samples'.format(task), "Directory name to save the image samples [samples]") flags.DEFINE_string("direction", "forward", "The direction of generator [forward, backward]")
flags.DEFINE_string("test_dir", "/home/liuwenjie/deep_save/{}/test".format(task), "The direction of test")
flags.DEFINE_boolean("is_train", True, "True for training, False for testing [False]")
flags.DEFINE_boolean("is_crop", False, "True for training, False for testing [False]")
# flags.DEFINE_boolean("visualize", False, "True for visualizing, False for nothing [False]")
FLAGS = flags.FLAGS
通过⼀个⽹络的学习，我发现⼤多数⽹络都需要进⾏以上的定义，以下⼀⼀解读: 1. task:(dcgan_spring2winter_18.3.1)设定任务名称,⼀般设计成为 {model name}-{data name}-{time}，这样每次训练都可以获得唯⼀确定的ckpt和image，保证多任务不重复
2. n_ epoch:(200)设定task整批数据集运⾏的轮次 3. learning_rate :(0.002)⼀般起始的learning rate 设定为0.01，在神经⽹络的loss停⽌不下降的时候,lr下降⼀个数量级 4. beta1 :(0.5)设定在0.9或0.5左右使⽤，主要是给Adam作为参数
5. weight_decay:(1e-5)下降权重,⼀般不设置
6. pool_size:(50)
7. train_size:(len<train_dataset)训练数据集,⼀般都⼩于训练集，训练时需要len(min(min(train_A,tran_label),train_size))防⽌错误
8. batch_size:(1-10)⼀次放⼊⽹络的图⽚数量
9. mf_dim:(64)⽹络的第⼀个卷积层使⽤的卷积核个数
10.output_size:(256)输出图⽚的数量
11.sample_size:(64)sample的图⽚的数量
12.c_dim:(3)图⽚的通道数⽬
13.sample_step:(500)运⾏多少步保存⼀次sample
14.save_step:(10)运⾏多少次进⾏⼀次step的保存⼯作
15.dataset_dir:(/home/deep_save/{}/datasetdir.format(task))⽂件存储位置
人耳鼠16.ckpt_dir:(/home/deep_save/{}/ckpt_dir.format(task))ckpt存储位置
17.sample_dir:(/home/deep_save/{}/sample_dir.format(task))sample存储的⽂件夹
18.direction::(forward,backword)正向传播，反向传播
20.is_train:train或test
21.is_crop：是否需要将输⼊剪裁
在使⽤机器学习的过程中，通常需要使⽤预训练模型安全心理学论文
最基础的⼏个预训练模型都是使⽤ImageNet进⾏训练的resnet,cgg16,googleNet.⽆论是做分类，分割或者是做检测问题，都需要使⽤这三种的⼀个，不然都很难获得⼀个较为良好的训练结果.以tensorlayer中使⽤vgg16为例，演⽰⼀下如何使⽤tensorlayer逐步对每个层进⾏参数加载:
#! /usr/bin/python
# -*- coding: utf-8 -*-
"""
VGG-16 for ImageNet.
Introduction
----------------
VGG is a convolutional neural network model proposed by K. Simonyan and A. Zisserman
from the University of Oxford in the paper “Very Deep Convolutional Networks for
Large-Scale Image Recognition” . The model achieves 92.7% top-5 test accuracy in ImageNet,
which is a dataset of over 14 million images belonging to 1000 classes.
Download Pre-trained Model
----------------------------
- Model weights in this example - vgg16_weights.npz : o.edu/~frossard/post/vgg16/
采集重构
- Caffe VGG 16 model : gist.github/ksimonyan/211839e770f7b538e2d8#file-readme-md
- Tool to convert the Caffe models to TensorFlow's : github/ethereon/caffe-tensorflow
Note
------
- For simplified CNN layer see "Convolutional layer (Simplified)"
in read the docs website.
- When feeding other images to the model be sure to properly resize or crop them
beforehand. Distorted images might end up being misclassified. One way of safely
feeding images of multiple sizes is by doing center cropping, as shown in the
following snippet:
>>> image_h, image_w, _ = np.shape(img)
>>> shorter_side = min(image_h, image_w)
>>> scale = 224. / shorter_side
>>> image_h, image_w = np.ceil([scale * image_h, scale * image_w]).astype('int32')
>>> img = imresize(img, (image_h, image_w))
>>> crop_x = (image_w - 224) / 2
>>> crop_y = (image_h - 224) / 2
>>> img = img[crop_y:crop_y+224,crop_x:crop_x+224,:]
"""
import os
import time
import numpy as np
from scipy.misc import imread, imresize
import tensorflow as tf
import tensorlayer as tl
from tensorlayer.layers import *
# try:
# dels.imagenet_classes import *
# except Exception as e:
# raise Exception(
# "{} / download the file from: github/zsdonghao/tensorlayer/tree/master/example/data".format(e)
# )
def conv_layers(net_in):
with tf.name_scope('preprocess'):
# Notice that we include a preprocessing layer that takes the RGB image
# with pixels values in the range of 0-255 and subtracts the mean image
# values (calculated over the entire ImageNet training set).
mean = tf.constant([123.68, 116.779, 103.939], dtype=tf.float32, shape=[1, 1, 1, 3], name='img_mean')
net_in.outputs = net_in.outputs - mean
# conv1
net = Conv2dLayer(net_in, lu, shape=[3, 3, 3, 64], strides=[1, 1, 1, 1], padding='SAME', name='conv1_1') net = Conv2dLayer(net, lu, shape=[3, 3, 64, 64], strides=[1, 1, 1, 1], padding='SAME', name='conv1_2') net = PoolLayer(net, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME', ax_pool, name='pool1')
# conv2
net = Conv2dLayer(net, lu, shape=[3, 3, 64, 128], strides=[1, 1, 1, 1], padding='SAME', name='conv2_1') net = Conv2dLayer(net, lu, shape=[3, 3, 128, 128], strides=[1, 1, 1, 1], padding='SAME', name='conv2_2') net = PoolLayer(net, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME', ax_pool, name='pool2')
# conv3
net = Conv2dLayer(net, lu, shape=[3, 3, 128, 256], strides=[1, 1, 1, 1], padding='SAME', name='conv3_1') net = Conv2dLayer(net, lu, shape=[3, 3, 256, 256], strides=[1, 1, 1, 1], padding='SAME', name='conv3_2') net = Conv2dLayer(net, lu, shape=[3, 3, 256, 256], strides=[1, 1, 1, 1], padding='SAME', name='conv3_3') net = PoolLayer(net, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME', ax_pool, name='pool3')
# conv4
net = Conv2dLayer(net, lu, shape=[3, 3, 256, 512], strides=[1, 1, 1, 1], padding='SAME', name='conv4_1') net = Conv2dLayer(net, lu, shape=[3, 3, 512, 512], strides=[1, 1, 1, 1], padding='SAME', name='conv4_2') net = Conv2dLayer(net, lu, shape=[3, 3, 512, 512], strides=[1, 1, 1, 1], padding='SAME', name='conv4_3') net = PoolLayer(net, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME', ax_pool, name='pool4')
# conv5
net = Conv2dLayer(net, lu, shape=[3, 3, 512, 512], strides=[1, 1, 1, 1], padding='SAME', name='conv5_1') net = Conv2dLayer(net, lu, shape=[3, 3, 512, 512], strides=[1, 1, 1, 1], padding='SAME', name='conv5_2') net = Conv2dLayer(net, lu, shape=[3, 3, 512, 51
2], strides=[1, 1, 1, 1], padding='SAME', name='conv5_3') net = PoolLayer(
net, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME', ax_pool, name='pool5')
return net
def conv_layers_simple_api(net_in):
with tf.name_scope('preprocess'):
# Notice that we include a preprocessing layer that takes the RGB image
# with pixels values in the range of 0-255 and subtracts the mean image
# values (calculated over the entire ImageNet training set).
mean = tf.constant([123.68, 116.779, 103.939], dtype=tf.float32, shape=[1, 1, 1, 3], name='img_mean')
net_in.outputs = net_in.outputs - mean
# conv1
net = Conv2d(net_in, 64, filter_size=(3, 3), strides=(1, 1), lu, padding='SAME', name='conv1_1')
net = Conv2d(net, n_filter=64, filter_size=(3, 3), strides=(1, 1), lu, padding='SAME', name='conv1_2')
net = MaxPool2d(net, filter_size=(2, 2), strides=(2, 2), padding='SAME', name='pool1')
# conv2
net = Conv2d(net, n_filter=128, filter_size=(3, 3), strides=(1, 1), lu, padding='SAME', name='conv2_1')
net = Conv2d(net, n_filter=128, filter_size=(3, 3), strides=(1, 1), lu, padding='SAME', name='conv2_2')
net = MaxPool2d(net, filter_size=(2, 2), strides=(2, 2), padding='SAME', name='pool2')
# conv3
net = Conv2d(net, n_filter=256, filter_size=(3, 3), strides=(1, 1), lu, padding='SAME', name='conv3_1')
net = Conv2d(net, n_filter=256, filter_size=(3, 3), strides=(1, 1), lu, padding='SAME', name='conv3_2')
net = Conv2d(net, n_filter=256, filter_size=(3, 3), strides=(1, 1), lu, padding='SAME', name='conv3_3')
net = MaxPool2d(net, filter_size=(2, 2), strides=(2, 2), padding='SAME', name='pool3')
# conv4
net = Conv2d(net, n_filter=512, filter_size=(3, 3), strides=(1, 1), lu, padding='SAME', name='conv4_1')
net = Conv2d(net, n_filter=512, filter_size=(3, 3), strides=(1, 1), lu, padding='SAME', name='conv4_2')
net = Conv2d(net, n_filter=512, filter_size=(3, 3), strides=(1, 1), lu, padding='SAME', name='conv4_3')
net = MaxPool2d(net, filter_size=(2, 2), strides=(2, 2), padding='SAME', name='pool4')
# conv5
net = Conv2d(net, n_filter=512, filter_size=(3, 3), strides=(1, 1), lu, padding='SAME', name='conv5_1')
net = Conv2d(net, n_filter=512, filter_size=(3, 3), strides=(1, 1), lu, padding='SAME', name='conv5_2')
net = Conv2d(net, n_filter=512, filter_size=(3, 3), strides=(1, 1), lu, padding='SAME', name='conv5_3')
net = MaxPool2d(net, filter_size=(2, 2), strides=(2, 2), padding='SAME', name='pool5')
return net
def fc_layers(net):
net = FlattenLayer(net, name='flatten')
net = DenseLayer(net, n_units=4096, lu, name='fc1_relu')
net = DenseLayer(net, n_units=4096, lu, name='fc2_relu')
net = DenseLayer(net, n_units=1000, lu, name='fc3_relu')
return net
sess = tf.InteractiveSession()
x = tf.placeholder(tf.float32, [None, 224, 224, 3])
# y_ = tf.placeholder(tf.int32, shape=[None, ], name='y_')
net_in = InputLayer(x, name='input')
# net_cnn = conv_layers(net_in) # professional CNN APIs
net_cnn = conv_layers_simple_api(net_in) # simplified CNN APIs
net = fc_layers(net_cnn)
y = net.outputs
probs = tf.nn.softmax(y)
y1 = net_cnn.outputs
# y_op = tf.softmax(y), 1)
# cost = ss_entropy(y, y_, name='cost')
# correct_prediction = tf.equal(tf.cast(tf.argmax(y, 1), tf.float32), tf.cast(y_, tf.float32))
# acc = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
tl.layers.initialize_global_variables(sess)
net.print_params()
net.print_layers()
tl.files.maybe_download_and_extract(
'vgg16_weights.npz', '/home/liuwenjie/premodels', 'o.edu/~frossard/vgg16/', expected_bytes=553436134 )
大型圆形水池模板施工视频
npz = np.load(os.path.join('/home/liuwenjie/premodels', 'vgg16_weights.npz'))
params = []
for val in sorted(npz.items()):
print(" Loading params %s" % str(val[1].shape))
params.append(val[1])
tl.files.assign_params(sess, params[0:16], net_cnn)
img1 = imread('/home/liuwenjie/liuwenjie/tensorflow_workplace/image2012_1.jpg', mode='RGB') # test data in github
img1 = imresize(img1, (224, 224))
y = sess.run(probs, feed_dict={x: [img1]})[0] # 1st time take time to compile
江苏公路信息网
# start_time = time.time()
# prob = sess.run(probs, feed_dict={x: [img1]})[0]
# print(" End time : %.5ss" % (time.time() - start_time))
# preds = (np.argsort(prob)[::-1])[0:5]
# for p in preds:
# print(p, prob[p])
ys = sess.run(y1,feed_dict={x:[img1]})[0]
print(y)
print(ys)
在这⾥我复制了⽹上vgg16的参数加载代码，可以看到，在预处理阶段，使⽤减去均值的办法进⾏处理，将图⽚输⼊，若只需要卷积层不需要全连接层，则将原来npz的前16层的权重加⼊net_cnn，因为全连接层的数据占全部参数的70%以上，所以做语义分割的过程中不是很建议使⽤denselayer层的参
数.在使⽤了这⼀系列操作之后，使⽤⼀张图⽚测试⼀下带denselayer的检测图⽚的结果，若有显⽰则加载预训练模型成功
豫公网安备41160202000603
站长QQ:729038198
关于我们
投诉建议