caffe如何自定义网络以及自定义层（python）（三）

前面是学习调用和修改训练好的model，接下来是训练自己的模型，这个其实很简单了，前面已经介绍过很多种数据输入格式了，直接原来的配置文件中的数据替换成自己的数据。这一篇学习一下LeNet的python求解，按照官网的整理一下。

首先是数据集和配置文件准备：

caffe_root=/home/x/git/caffe
pro=/home/x/git/caffe/examples/test_layer_lb/test_python/mnist
sh $caffe_root/data/mnist/get_mnist.sh
sh $caffe_root/examples/mnist/create_mnist.sh

调用python接口的调试代码：

#!/usr/bin/env python
#coding=utf-8
import numpy as np
from pylab import *
caffe_root = '/home/x/git/caffe/'
import sys
sys.path.insert(0, caffe_root + 'python')
import caffe
# run scripts from caffe root
import os
os.chdir(caffe_root)
os.chdir('examples')
from caffe import layers as L, params as P
#生成配置文件
def lenet(lmdb, batch_size):
    # our version of LeNet: a series of linear and simple nonlinear transformations
    #定义网络
    n = caffe.NetSpec()
    
    n.data, n.label = L.Data(batch_size=batch_size, backend=P.Data.LMDB, source=lmdb,
                             transform_param=dict(scale=1./255), ntop=2)
    n.conv1 = L.Convolution(n.data, kernel_size=5, num_output=20, weight_filler=dict(type='xavier'))
    n.pool1 = L.Pooling(n.conv1, kernel_size=2, stride=2, pool=P.Pooling.MAX)
    n.conv2 = L.Convolution(n.pool1, kernel_size=5, num_output=50, weight_filler=dict(type='xavier'))
    n.pool2 = L.Pooling(n.conv2, kernel_size=2, stride=2, pool=P.Pooling.MAX)
    n.fc1 =   L.InnerProduct(n.pool2, num_output=500, weight_filler=dict(type='xavier'))
    n.relu1 = L.ReLU(n.fc1, in_place=True)
    n.score = L.InnerProduct(n.relu1, num_output=10, weight_filler=dict(type='xavier'))
    n.loss =  L.SoftmaxWithLoss(n.score, n.label)
    return n.to_proto()
    
with open('/home/x/git/caffe/examples/mnist/lenet_auto_train.prototxt', 'w') as f:
    f.write(str(lenet('/home/x/git/caffe/examples/mnist/mnist_train_lmdb', 64)))
    
with open('/home/x/git/caffe/examples/mnist/lenet_auto_test.prototxt', 'w') as f:
    f.write(str(lenet('/home/x/git/caffe/examples/mnist/mnist_test_lmdb', 100)))
#加载求解文件和调试solver文件
caffe.set_device(0)
caffe.set_mode_gpu()

### load the solver and create train and test nets
solver = None  # ignore this workaround for lmdb data (can't instantiate two solvers on the same data)
solver = caffe.SGDSolver('/home/x/git/caffe/examples/mnist/lenet_auto_solver.prototxt')

# each output is (batch size, feature dim, spatial dim)
print [(k, v.data.shape) for k, v in solver.net.blobs.items()]
# just print the weight sizes (we'll omit the biases)
print [(k, v[0].data.shape) for k, v in solver.net.params.items()]
print "++++++++++++++++++++++++++++++++++++net+++++++++++++++++++++++++++++++++++++++++"
print type(solver)
print dir(solver)
#print help(solver)
#print help(solver.net)
#forward是个字典
print dir(solver.net.forward())
solver.net.forward()  # train net
print "++++++++++++++++++++++++++++++++test_nets+++++++++++++++++++++++++++++++++++++++"
print dir(solver.test_nets)
print len(solver.test_nets)
print type(solver.test_nets)
#是个长度为1的向量
print dir(solver.test_nets[0])
#是个实例化的方法
print type(solver.test_nets[0].forward)
print "++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++"
solver.test_nets[0].forward()  # test net (there can be more than one)


# we use a little trick to tile the first eight images
imshow(solver.net.blobs['data'].data[:8, 0].transpose(1, 0, 2).reshape(28, 8*28), cmap='gray'); axis('off')
print 'train labels:', solver.net.blobs['label'].data[:8]
#show()
imshow(solver.test_nets[0].blobs['data'].data[:8, 0].transpose(1, 0, 2).reshape(28, 8*28), cmap='gray'); axis('off')
print 'test labels:', solver.test_nets[0].blobs['label'].data[:8]
#show()
print "++++++++++++++++++++++++++++++++++step(1)++++++++++++++++ ++++++++++++++++++++++"
print dir(solver.step)
print type(solver.step)
#iteration 0，不计入
#solver.step(1)
print "++++++++++++++++++++++++++++++++++step(2)+++++++++++++++++++++++++++++++++++++++"
#这个是迭代iteration个数，然后显示设置在配置文件中我设置1，5，10次(默认100)
solver.step(40)
print "++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++"
imshow(solver.net.params['conv1'][0].diff[:, 0].reshape(4, 5, 5, 5)
       .transpose(0, 2, 1, 3).reshape(4*5, 5*5), cmap='gray'); axis('off')
#show()
#一些求解迭代参数设置
#由于前面有迭代40次所以最后显示迭代220次(从0开始)
niter = 200
test_interval = 25
# losses will also be stored in the log
train_loss = zeros(niter)
test_acc = zeros(int(np.ceil(niter / test_interval)))
output = zeros((niter, 8, 10))

# the main solver loop
for it in range(niter):
    solver.step(1)  # SGD by Caffe
    # store the train loss
    train_loss[it] = solver.net.blobs['loss'].data
    
    # store the output on the first test batch
    # (start the forward pass at conv1 to avoid loading new data)
    solver.test_nets[0].forward(start='conv1')
    output[it] = solver.test_nets[0].blobs['score'].data[:8]

    # run a full test every so often
    # (Caffe can also do this for us and write to a log, but we show here
    #  how to do it directly in Python, where more complicated things are easier.)
    if it % test_interval == 0:
        print 'Iteration', it, 'testing...'
        correct = 0
        for test_it in range(100):
            solver.test_nets[0].forward()
            correct += sum(solver.test_nets[0].blobs['score'].data.argmax(1)
                           == solver.test_nets[0].blobs['label'].data)
        test_acc[it // test_interval] = correct / 1e4
#plot the train loss and test accuracy.
_, ax1 = subplots()
ax2 = ax1.twinx()
ax1.plot(arange(niter), train_loss)
ax2.plot(test_interval * arange(len(test_acc)), test_acc, 'r')
ax1.set_xlabel('iteration')
ax1.set_ylabel('train loss')
ax2.set_ylabel('test accuracy')
ax2.set_title('Test Accuracy: {:.2f}'.format(test_acc[-1]))
show()

for i in range(8):
    figure(figsize=(2, 2))
    imshow(solver.test_nets[0].blobs['data'].data[i, 0], cmap='gray')
    figure(figsize=(10, 2))
    imshow(output[:50, i].T, interpolation='nearest', cmap='gray')
    xlabel('iteration')
    ylabel('label')
    #show()

for i in range(8):
    figure(figsize=(2, 2))
    imshow(solver.test_nets[0].blobs['data'].data[i, 0], cmap='gray')
    figure(figsize=(10, 2))
    imshow(exp(output[:50, i].T) / exp(output[:50, i].T).sum(0), interpolation='nearest', cmap='gray')
    xlabel('iteration')
    ylabel('label')
    #show()

最后的运行版本：

#!/usr/bin/env python
#coding=utf-8
import numpy as np
from pylab import *
caffe_root = '/home/x/git/caffe/'
import sys
sys.path.insert(0, caffe_root + 'python')
import caffe
# run scripts from caffe root
import os
os.chdir(caffe_root)
os.chdir('examples')
from caffe import layers as L, params as P

train_net_path = '/home/x/git/caffe/examples/mnist/custom_auto_train.prototxt'
test_net_path = '/home/x/git/caffe/examples/mnist/custom_auto_test.prototxt'
solver_config_path = '/home/x/git/caffe/examples/mnist/custom_auto_solver.prototxt'

### define net
def custom_net(lmdb, batch_size):
    # define your own net!
    n = caffe.NetSpec()
    
    # keep this data layer for all networks
    n.data, n.label = L.Data(batch_size=batch_size, backend=P.Data.LMDB, source=lmdb,
                             transform_param=dict(scale=1./255), ntop=2)
    
    # EDIT HERE to try different networks
    # this single layer defines a simple linear classifier
    # (in particular this defines a multiway logistic regression)
    n.score =   L.InnerProduct(n.data, num_output=10, weight_filler=dict(type='xavier'))
    
    # EDIT HERE this is the LeNet variant we have already tried
    # n.conv1 = L.Convolution(n.data, kernel_size=5, num_output=20, weight_filler=dict(type='xavier'))
    # n.pool1 = L.Pooling(n.conv1, kernel_size=2, stride=2, pool=P.Pooling.MAX)
    # n.conv2 = L.Convolution(n.pool1, kernel_size=5, num_output=50, weight_filler=dict(type='xavier'))
    # n.pool2 = L.Pooling(n.conv2, kernel_size=2, stride=2, pool=P.Pooling.MAX)
    # n.fc1 =   L.InnerProduct(n.pool2, num_output=500, weight_filler=dict(type='xavier'))
    # EDIT HERE consider L.ELU or L.Sigmoid for the nonlinearity
    # n.relu1 = L.ReLU(n.fc1, in_place=True)
    # n.score =   L.InnerProduct(n.fc1, num_output=10, weight_filler=dict(type='xavier'))
    
    # keep this loss layer for all networks
    n.loss =  L.SoftmaxWithLoss(n.score, n.label)
    
    return n.to_proto()

with open(train_net_path, 'w') as f:
    f.write(str(custom_net('mnist/mnist_train_lmdb', 64)))    
with open(test_net_path, 'w') as f:
    f.write(str(custom_net('mnist/mnist_test_lmdb', 100)))

### define solver
from caffe.proto import caffe_pb2
s = caffe_pb2.SolverParameter()

# Set a seed for reproducible experiments:
# this controls for randomization in training.
s.random_seed = 0xCAFFE

# Specify locations of the train and (maybe) test networks.
s.train_net = train_net_path
s.test_net.append(test_net_path)
s.test_interval = 500  # Test after every 500 training iterations.
s.test_iter.append(100) # Test on 100 batches each time we test.

s.max_iter = 10000     # no. of times to update the net (training iterations)
 
# EDIT HERE to try different solvers
# solver types include "SGD", "Adam", and "Nesterov" among others.
s.type = "SGD"

# Set the initial learning rate for SGD.
s.base_lr = 0.01  # EDIT HERE to try different learning rates
# Set momentum to accelerate learning by
# taking weighted average of current and previous updates.
s.momentum = 0.9
# Set weight decay to regularize and prevent overfitting
s.weight_decay = 5e-4

# Set `lr_policy` to define how the learning rate changes during training.
# This is the same policy as our default LeNet.
s.lr_policy = 'inv'
s.gamma = 0.0001
s.power = 0.75
# EDIT HERE to try the fixed rate (and compare with adaptive solvers)
# `fixed` is the simplest policy that keeps the learning rate constant.
# s.lr_policy = 'fixed'

# Display the current training loss and accuracy every 1000 iterations.
s.display = 1000

# Snapshots are files used to store networks we've trained.
# We'll snapshot every 5K iterations -- twice during training.
s.snapshot = 5000
s.snapshot_prefix = '/home/x/git/caffe/examples/test_layer_lb/test_python/mnist/custom_net'

# Train on the GPU
s.solver_mode = caffe_pb2.SolverParameter.GPU

# Write the solver to a temporary file and return its filename.
with open(solver_config_path, 'w') as f:
    f.write(str(s))

### load the solver and create train and test nets
solver = None  # ignore this workaround for lmdb data (can't instantiate two solvers on the same data)
solver = caffe.get_solver(solver_config_path)

### solve
niter = 2500  # EDIT HERE increase to train for longer
test_interval = niter / 10
# losses will also be stored in the log
train_loss = zeros(niter)
test_acc = zeros(int(np.ceil(niter / test_interval)))

# the main solver loop
for it in range(niter):
    solver.step(1)  # SGD by Caffe
    
    # store the train loss
    train_loss[it] = solver.net.blobs['loss'].data
    
    # run a full test every so often
    # (Caffe can also do this for us and write to a log, but we show here
    #  how to do it directly in Python, where more complicated things are easier.)
    if it % test_interval == 0:
        print 'Iteration', it, 'testing...'
        correct = 0
        for test_it in range(100):
            solver.test_nets[0].forward()
            correct += sum(solver.test_nets[0].blobs['score'].data.argmax(1)
                           == solver.test_nets[0].blobs['label'].data)
        test_acc[it // test_interval] = correct / 1e4

_, ax1 = subplots()
ax2 = ax1.twinx()
ax1.plot(arange(niter), train_loss)
ax2.plot(test_interval * arange(len(test_acc)), test_acc, 'r')
ax1.set_xlabel('iteration')
ax1.set_ylabel('train loss')
ax2.set_ylabel('test accuracy')
ax2.set_title('Custom Test Accuracy: {:.2f}'.format(test_acc[-1]))
show()