def get_kaggle_reader(validation_pct=.2):
train_file = Path('./train.zip')
test_file = Path('./test.zip')
if not train_file.is_file() or not test_file.is_file():
print('👇👇👇👇👇👇👇👇👇👇👇👇👇')
print(
'Please put train.zip and test.zip from '
'https://www.kaggle.com/c/dogs-vs-cats-redux-kernels-edition to project root path.'
)
raise FileNotFoundError()
train_arc = ZipTarReader.auto(train_file.name)
test_arc = ZipTarReader.auto(test_file.name)
file_list = train_arc.shuffled_namelist()
file_list = np.array(file_list)
mid = int(len(file_list) * validation_pct)
train_file_list = file_list[mid:]
vali_file_list = file_list[:mid]
return BatchReader.BatchReader(train_arc, train_file_list, KaggleLabelClassifier),\
BatchReader.BatchReader(train_arc, vali_file_list, KaggleLabelClassifier),\
BatchReader.BatchReader(test_arc, test_arc.namelist(), KaggleLabelClassifier)
def main(args):
batch_size = 6
if FLAGS.mode == 'decode':
batch_size = FLAGS.beam_size
hps = ABS.HParams(
batch_size=batch_size,
mode=FLAGS.mode,
num_softmax_samples=4056,
min_lr=0.01,
lr=0.9,
hid_dim=400,
emb_dim=200,
max_grad_norm=2,
min_input_len=2, # discard articles/summaries < than this
C=5,
Q=2,
dec_timesteps=90,
enc_timesteps=520)
vocab = Data.Vocab(FLAGS.vocab_path)
batcher = BatchReader.Batcher(
vocab,
hps,
max_article_sentences=FLAGS.max_article_sentences,
max_abstract_sentences=FLAGS.max_abstract_sentences,
)
vocab_size = vocab.NumIds()
pad_id = vocab.WordToId(Data.PAD_TOKEN)
tf.set_random_seed(FLAGS.random_seed)
if hps.mode == 'train':
model = ABS.ABS(hps, pad_id, vocab_size)
_Train(model, batcher)
elif hps.mode == 'decode':
newhps = hps._replace(dec_timesteps=hps.C)
model = ABS.ABS(hps, pad_id, vocab_size)
decoder = Decode.BSDecoder(model, batcher, newhps, vocab)
decoder.DecodeLoop()
def main(argv=None):
# define placeholder{keep_probability,image,annotation}
keep_probability = tf.placeholder(tf.float32, name="keep_probabilty")
image = tf.placeholder(tf.float32,
shape=[None, IMAGE_SIZE, IMAGE_SIZE, 3],
name="input_image")
annotation = tf.placeholder(tf.int32,
shape=[None, IMAGE_SIZE, IMAGE_SIZE, 1],
name="annotation")
# execute inference then get pred_annotation, logits, and loss
pred_annotation, logits = inference(image, keep_probability)
tf.summary.image("input_image", image, max_outputs=2)
tf.summary.image("ground_truth",
tf.cast(annotation, tf.uint8),
max_outputs=2)
tf.summary.image("pred_annotation",
tf.cast(pred_annotation, tf.uint8),
max_outputs=2)
loss = tf.reduce_mean((tf.nn.sparse_softmax_cross_entropy_with_logits(
logits=logits,
labels=tf.squeeze(annotation, squeeze_dims=[3]),
name="entropy")))
tf.summary.scalar("entropy", loss)
trainable_var = tf.trainable_variables()
train_op = train(loss, trainable_var)
print("Setting up summary op...")
summary_op = tf.summary.merge_all()
print("Setting up image reader...")
# Data_zoo/MIT_SceneParsing/
train_records, valid_records = scene_parsing.read_dataset(FLAGS.data_dir)
print(len(train_records))
print(len(valid_records))
print("Setting up dataset reader")
image_options = {'resize': True, 'resize_size': IMAGE_SIZE}
if FLAGS.mode == 'train':
train_dataset_reader = dataset.BatchDatset(train_records,
image_options)
validation_dataset_reader = dataset.BatchDatset(valid_records,
image_options)
sess = tf.Session()
print("Setting up Saver...")
saver = tf.train.Saver()
summary_writer = tf.summary.FileWriter(FLAGS.logs_dir, sess.graph)
sess.run(tf.global_variables_initializer())
ckpt = tf.train.get_checkpoint_state(FLAGS.logs_dir)
if ckpt and ckpt.model_checkpoint_path:
saver.restore(sess, ckpt.model_checkpoint_path)
print("Model restored...")
if FLAGS.mode == "train":
for itr in range(MAX_ITERATION):
train_images, train_annotations = train_dataset_reader.next_batch(
FLAGS.batch_size)
feed_dict = {
image: train_images,
annotation: train_annotations,
keep_probability: 0.85
}
sess.run(train_op, feed_dict=feed_dict)
if itr % 10 == 0:
train_loss, summary_str = sess.run([loss, summary_op],
feed_dict=feed_dict)
print("Step: %d, Train_loss:%g" % (itr, train_loss))
summary_writer.add_summary(summary_str, itr)
if itr % 500 == 0:
valid_images, valid_annotations = validation_dataset_reader.next_batch(
FLAGS.batch_size)
valid_loss = sess.run(loss,
feed_dict={
image: valid_images,
annotation: valid_annotations,
keep_probability: 1.0
})
print("%s ---> Validation_loss: %g" %
(datetime.datetime.now(), valid_loss))
saver.save(sess, FLAGS.logs_dir + "model.ckpt", itr)
elif FLAGS.mode == "visualize":
valid_images, valid_annotations = validation_dataset_reader.get_random_batch(
FLAGS.batch_size)
pred = sess.run(pred_annotation,
feed_dict={
image: valid_images,
annotation: valid_annotations,
keep_probability: 1.0
})
# single channle
valid_annotations = np.squeeze(valid_annotations, axis=3)
pred = np.squeeze(pred, axis=3)
for itr in range(FLAGS.batch_size):
utils.save_image(valid_images[itr].astype(np.uint8),
FLAGS.logs_dir,
name="inp_" + str(5 + itr))
utils.save_image(valid_annotations[itr].astype(np.uint8),
FLAGS.logs_dir,
name="gt_" + str(5 + itr))
utils.save_image(pred[itr].astype(np.uint8),
FLAGS.logs_dir,
name="pred_" + str(5 + itr))
print("Saved image: %d" % itr)
def evaluate_lenet5(learning_rate=0.1, n_epochs=200,
dataset='mnist.pkl.gz',
nkerns=[20, 50], batch_size=500):
""" Demonstrates lenet on MNIST dataset
:type learning_rate: float
:param learning_rate: learning rate used (factor for the stochastic
gradient)
:type n_epochs: int
:param n_epochs: maximal number of epochs to run the optimizer
:type dataset: string
:param dataset: path to the dataset used for training /testing (MNIST here)
:type nkerns: list of ints
:param nkerns: number of kernels on each layer
"""
rng = numpy.random.RandomState(23455)
br = BatchReader.inputs()
br2 = BatchReader.inputs(testingData = True)
X, Y = br.getNPArray(2)
train_set_x = theano.shared(X[0:50000*0.6].astype(float))
train_set_y = theano.shared(Y[0:50000*0.6].astype('int32'))
valid_set_x = theano.shared(X[50000*0.6:50000*0.9].astype(float))
valid_set_y = theano.shared(Y[50000*0.6:50000*0.9].astype('int32'))
test_set_x = theano.shared(X[50000*0.9:50000].astype(float))
test_set_y = theano.shared(Y[50000*0.9:50000].astype('int32'))
#datasets = load_data(dataset)
#train_set_x, train_set_y = datasets[0]
#valid_set_x, valid_set_y = datasets[1]
#test_set_x, test_set_y = datasets[2]
# compute number of minibatches for training, validation and testing
n_train_batches = train_set_x.get_value(borrow=True).shape[0]
n_valid_batches = valid_set_x.get_value(borrow=True).shape[0]
n_test_batches = test_set_x.get_value(borrow=True).shape[0]
n_train_batches /= batch_size
n_valid_batches /= batch_size
n_test_batches /= batch_size
# allocate symbolic variables for the data
index = T.lscalar() # index to a [mini]batch
# start-snippet-1
x = T.matrix('x') # the data is presented as rasterized images
y = T.ivector('y') # the labels are presented as 1D vector of
# [int] labels
######################
# BUILD ACTUAL MODEL #
######################
print '... building the model'
# Reshape matrix of rasterized images of shape (batch_size, 28 * 28)
# to a 4D tensor, compatible with our LeNetConvPoolLayer
# (28, 28) is the size of MNIST images.
layer0_input = x.reshape((batch_size, 1, 48, 48))
# Construct the first convolutional pooling layer:
# filtering reduces the image size to (28-5+1 , 28-5+1) = (24, 24)
# maxpooling reduces this further to (24/2, 24/2) = (12, 12)
# 4D output tensor is thus of shape (batch_size, nkerns[0], 12, 12)
layer0 = LeNetConvPoolLayer(
rng,
input=layer0_input,
image_shape=(batch_size, 1, 48, 48),
filter_shape=(nkerns[0], 1, 9, 9),
poolsize=(2, 2)
)
# Construct the second convolutional pooling layer
# filtering reduces the image size to (12-5+1, 12-5+1) = (8, 8)
# maxpooling reduces this further to (8/2, 8/2) = (4, 4)
# 4D output tensor is thus of shape (batch_size, nkerns[1], 4, 4)
layer1 = LeNetConvPoolLayer(
rng,
input=layer0.output,
image_shape=(batch_size, nkerns[0], 20, 20),
filter_shape=(nkerns[1], nkerns[0], 8, 8),
poolsize=(2, 2)
)
# the HiddenLayer being fully-connected, it operates on 2D matrices of
# shape (batch_size, num_pixels) (i.e matrix of rasterized images).
# This will generate a matrix of shape (batch_size, nkerns[1] * 4 * 4),
# or (500, 50 * 4 * 4) = (500, 800) with the default values.
layer2_input = layer1.output.flatten(2)
# construct a fully-connected sigmoidal layer
layer2 = HiddenLayer(
rng,
input=layer2_input,
n_in=nkerns[1] * 6 * 6,
n_out=500,
activation=T.tanh
)
# classify the values of the fully-connected sigmoidal layer
layer3 = LogisticRegression(input=layer2.output, n_in=500, n_out=10)
# the cost we minimize during training is the NLL of the model
cost = layer3.negative_log_likelihood(y)
print type(test_set_x)
# create a function to compute the mistakes that are made by the model
test_model = theano.function(
[index],
layer3.errors(y),
givens={
x: test_set_x[index * batch_size: (index + 1) * batch_size],
y: test_set_y[index * batch_size: (index + 1) * batch_size]
}
)
validate_model = theano.function(
[index],
layer3.errors(y),
givens={
x: valid_set_x[index * batch_size: (index + 1) * batch_size],
y: valid_set_y[index * batch_size: (index + 1) * batch_size]
}
)
# create a list of all model parameters to be fit by gradient descent
params = layer3.params + layer2.params + layer1.params + layer0.params
# create a list of gradients for all model parameters
grads = T.grad(cost, params)
# train_model is a function that updates the model parameters by
# SGD Since this model has many parameters, it would be tedious to
# manually create an update rule for each model parameter. We thus
# create the updates list by automatically looping over all
# (params[i], grads[i]) pairs.
updates = [
(param_i, param_i - learning_rate * grad_i)
for param_i, grad_i in zip(params, grads)
]
train_model = theano.function(
[index],
cost,
updates=updates,
givens={
x: train_set_x[index * batch_size: (index + 1) * batch_size],
y: train_set_y[index * batch_size: (index + 1) * batch_size]
}
)
# end-snippet-1
###############
# TRAIN MODEL #
###############
print '... training'
# early-stopping parameters
patience = 10000 # look as this many examples regardless
patience_increase = 2 # wait this much longer when a new best is
# found
improvement_threshold = 0.995 # a relative improvement of this much is
# considered significant
validation_frequency = min(n_train_batches, patience / 2)
# go through this many
# minibatche before checking the network
# on the validation set; in this case we
# check every epoch
best_validation_loss = numpy.inf
best_iter = 0
test_score = 0.
start_time = timeit.default_timer()
epoch = 0
done_looping = False
while (epoch < n_epochs) and (not done_looping):
epoch = epoch + 1
for minibatch_index in xrange(n_train_batches):
iter = (epoch - 1) * n_train_batches + minibatch_index
if iter % 100 == 0:
print 'training @ iter = ', iter
cost_ij = train_model(minibatch_index)
if (iter + 1) % validation_frequency == 0:
# compute zero-one loss on validation set
validation_losses = [validate_model(i) for i
in xrange(n_valid_batches)]
this_validation_loss = numpy.mean(validation_losses)
print('epoch %i, minibatch %i/%i, validation error %f %%' %
(epoch, minibatch_index + 1, n_train_batches,
this_validation_loss * 100.))
# if we got the best validation score until now
if this_validation_loss < best_validation_loss:
#improve patience if loss improvement is good enough
if this_validation_loss < best_validation_loss * \
improvement_threshold:
patience = max(patience, iter * patience_increase)
# save best validation score and iteration number
best_validation_loss = this_validation_loss
best_iter = iter
# test it on the test set
test_losses = [
test_model(i)
for i in xrange(n_test_batches)
]
test_score = numpy.mean(test_losses)
print((' epoch %i, minibatch %i/%i, test error of '
'best model %f %%') %
(epoch, minibatch_index + 1, n_train_batches,
test_score * 100.))
if patience <= iter:
done_looping = True
break
end_time = timeit.default_timer()
print('Optimization complete.')
print('Best validation score of %f %% obtained at iteration %i, '
'with test performance %f %%' %
(best_validation_loss * 100., best_iter + 1, test_score * 100.))
print >> sys.stderr, ('The code for file ' +
os.path.split(__file__)[1] +
' ran for %.2fm' % ((end_time - start_time) / 60.))
print "Dumping parameters to ../data/convnet.pkl"
dir = os.getcwd()
path = os.path.join(dir,"../data")
os.chdir(path)
with open( "convnet.pkl" , 'wb') as file:
dict = {
"layer0" : layer0.param,
"layer1" : layer1.param,
"layer2" : layer2.param,
"layer3" : layer3.param
}
pickle.dump(dict, file, 2 )
os.chdir(dir)
import BatchReader as dataset
import tensorflow as tf
import numpy as np
max_it =int(1e4 + 1)
IMAGE_HEIGHT = 64
IMAGE_WIDTH = 256
MAX_CAPTCHA = 5
CHAR_SET_LEN = 10
batch_size = 64
file = open('acc.txt','a')
train_dataset_reader = dataset.BatchReader(index_file='index')
X = tf.placeholder(tf.float32, [None, IMAGE_HEIGHT*IMAGE_WIDTH])
Y = tf.placeholder(tf.float32, [None, MAX_CAPTCHA*CHAR_SET_LEN])
keep_prob = tf.placeholder(tf.float32) # dropout
# CNN
def crack_captcha_cnn(w_alpha=0.01, b_alpha=0.1):
x = tf.reshape(X, shape=[-1, IMAGE_HEIGHT, IMAGE_WIDTH, 1])
# 4 conv layer
w_c1 = tf.Variable(w_alpha*tf.random_normal([3, 3, 1, 32]))
b_c1 = tf.Variable(b_alpha*tf.random_normal([32]))
conv1 = tf.nn.relu(tf.nn.bias_add(tf.nn.conv2d(x, w_c1, strides=[1, 1, 1, 1], padding='SAME'), b_c1))
conv1 = tf.nn.max_pool(conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')
conv1 = tf.nn.dropout(conv1, keep_prob)
w_c2 = tf.Variable(w_alpha*tf.random_normal([3, 3, 32, 64]))
b_c2 = tf.Variable(b_alpha*tf.random_normal([64]))
conv2 = tf.nn.relu(tf.nn.bias_add(tf.nn.conv2d(conv1, w_c2, strides=[1, 1, 1, 1], padding='SAME'), b_c2))
def main(argv=None):
keep_probability = tf.placeholder(tf.float32, name="keep_probabilty")
learning_rate = tf.placeholder(tf.float32, name="learning_rate")
image = tf.placeholder(tf.float32, shape=[None, IMAGE_SIZE, IMAGE_SIZE, 7], name="input_image")
annotation = tf.placeholder(tf.int32, shape=[None, IMAGE_SIZE, IMAGE_SIZE, 1], name="annotation")
pred_annotation_value, pred_annotation, logits = inference(image, keep_probability)
#logits:the last layer of conv net
#labels:the ground truth
at = float(cfgs.at)
a_w = (1- 2*at) * tf.cast(tf.squeeze(annotation, squeeze_dims=[3]), tf.float32) + at
pro = tf.nn.softmax(logits)
loss_weight = tf.pow(1-tf.reduce_sum(pro * tf.one_hot(tf.squeeze(annotation, squeeze_dims=[3]), cfgs.NUM_OF_CLASSESS), 3), cfgs.gamma)
loss = tf.reduce_mean(loss_weight * a_w * tf.nn.sparse_softmax_cross_entropy_with_logits(logits=logits,
labels=tf.squeeze(annotation, squeeze_dims=[3]),
name="entropy"))
valid_loss = tf.reduce_mean((tf.nn.sparse_softmax_cross_entropy_with_logits(logits=logits,
labels=tf.squeeze(annotation, squeeze_dims=[3]),
name="valid_entropy")))
tf.summary.scalar("train_loss", loss)
tf.summary.scalar("valid_loss", valid_loss)
tf.summary.scalar("learning_rate", learning_rate)
trainable_var = tf.trainable_variables()
if cfgs.debug:
for var in trainable_var:
utils.add_to_regularization_and_summary(var)
train_op = train(loss, trainable_var, learning_rate)
print("Setting up summary op...")
summary_op = tf.summary.merge_all()
#Check if has the log file
if not os.path.exists(cfgs.logs_dir):
print("The logs path '%s' is not found" % cfgs.logs_dir)
print("Create now..")
os.makedirs(cfgs.logs_dir)
print("%s is created successfully!" % cfgs.logs_dir)
#Create a file to write logs.
#filename='logs'+ cfgs.mode + str(datatime.datatime.now()) + '.txt'
filename="logs_%s%s.txt"%(cfgs.mode,datetime.datetime.now())
path_=os.path.join(cfgs.logs_dir,filename)
with open(path_,'w') as logs_file:
logs_file.write("The logs file is created at %s.\n" % datetime.datetime.now())
logs_file.write("The model is ---%s---.\n" % cfgs.logs_dir)
logs_file.write("The mode is %s\n"% (cfgs.mode))
logs_file.write("The train data batch size is %d and the validation batch size is %d\n."%(cfgs.batch_size,cfgs.v_batch_size))
logs_file.write("The data size is %d and the MAX_ITERATION is %d.\n" % (IMAGE_SIZE, MAX_ITERATION))
logs_file.write("Setting up image reader...")
print("Setting up image reader...")
train_records, valid_records = scene_parsing.my_read_dataset(cfgs.seq_list_path, cfgs.anno_path)
print('number of train_records',len(train_records))
print('number of valid_records',len(valid_records))
with open(path_, 'a') as logs_file:
logs_file.write('number of train_records %d\n' % len(train_records))
logs_file.write('number of valid_records %d\n' % len(valid_records))
path_lr = cfgs.learning_rate_path
with open(path_lr, 'r') as f:
lr_ = float(f.readline().split('\n')[0])
print("Setting up dataset reader")
vis = True if cfgs.mode == 'all_visualize' else False
image_options = {'resize': True, 'resize_size': IMAGE_SIZE, 'visualize': vis, 'annotation':True}
if cfgs.mode == 'train':
train_dataset_reader = dataset.BatchDatset(train_records, image_options)
validation_dataset_reader = dataset.BatchDatset(valid_records, image_options)
#save current train itr from stopping(init = 0)
current_itr_var = tf.Variable(0, dtype=tf.int32, name='current_itr')
sess = tf.Session()
print("Setting up Saver...")
saver = tf.train.Saver()
summary_writer = tf.summary.FileWriter(cfgs.logs_dir, sess.graph)
sess.run(tf.global_variables_initializer())
ckpt = tf.train.get_checkpoint_state(cfgs.logs_dir)
#if not train,restore the model trained before
if ckpt and ckpt.model_checkpoint_path:
saver.restore(sess, ckpt.model_checkpoint_path)
print("Model restored...")
num_=0
current_itr = current_itr_var.eval(sess)
print('itr\tmode\tlr_\tacc\tiou_acc\tloss')
time_begin = time.time()
valid_time = 100
for itr in xrange(current_itr, MAX_ITERATION):
with open(path_lr, 'r') as f:
lr_ = float(f.readline().split('\n')[0])
train_images, train_annotations, if_finish, epoch_no= train_dataset_reader.next_batch(cfgs.batch_size)
if if_finish:
print('Epochs%d finished, time comsumed:%.5f' % (epoch_no, time.time()-time_begin))
time_begin = time.time()
feed_dict = {image: train_images, annotation: train_annotations, keep_probability: 0.85, learning_rate: lr_}
sess.run(train_op, feed_dict=feed_dict)
logs_file = open(path_, 'a')
if itr % 10 == 0:
train_loss, summary_str = sess.run([loss, summary_op], feed_dict=feed_dict)
print('%d\t%s\t%g\t%s\t%s\t%.6f' % (itr, 'train',lr_, 'None','None',train_loss))
summary_writer.add_summary(summary_str, itr)
if itr % valid_time == 0:
with open('time_to_valid.txt', 'r') as f:
valid_time = int(f.readline().split('\n')[0])
#Caculate the accurary at the training set.
train_random_images, train_random_annotations = train_dataset_reader.get_random_batch_for_train(cfgs.v_batch_size)
train_loss,train_pred_anno = sess.run([loss,pred_annotation], feed_dict={image:train_random_images,
annotation:train_random_annotations,
keep_probability:1.0})
accu_iou_,accu_pixel_ = accu.caculate_accurary(train_pred_anno, train_random_annotations)
print('%d\t%s\t%g\t%.5f\t%.5f\t%.6f' % (itr, 'train', lr_, accu_pixel_, accu_iou_, train_loss))
#Output the logs.
num_ = num_ + 1
logs_file.write('%d\t%s\t%g\t%.5f\t%.5f\t%.6f\n' % (itr, 'train', lr_, accu_pixel_, accu_iou_, train_loss))
valid_images, valid_annotations, _, _= validation_dataset_reader.next_batch(cfgs.v_batch_size)
#valid_loss = sess.run(loss, feed_dict={image: valid_images, annotation: valid_annotations,
#keep_probability: 1.0})
valid_loss_,pred_anno=sess.run([loss,pred_annotation],feed_dict={image:valid_images,
annotation:valid_annotations,
keep_probability:1.0})
accu_iou,accu_pixel=accu.caculate_accurary(pred_anno,valid_annotations)
print('%d\t%s\t%g\t%.5f\t%.5f\t%.6f' % (itr, 'valid', lr_, accu_pixel, accu_iou, valid_loss_))
#Output the logs.
logs_file.write('%d\t%s\t%g\t%.5f\t%.5f\t%.6f\n' % (itr, 'train', lr_, accu_pixel, accu_iou, valid_loss_))
#record current itr
sess.run(tf.assign(current_itr_var, itr))
saver.save(sess, cfgs.logs_dir + "model.ckpt", itr)
#End the iterator
'''
if itr % 10000 == 0 and itr > 0:
lr_ = lr_/10
print('learning rate change from %g to %g' %(lr_*10, lr_))
logs_file.write('learning rate change from %g to %g\n' %(lr_*10, lr_))'''
logs_file.close()
def get_data_video(self):
with tf.device('/cpu:0'):
self.video_dataset_reader = dataset.BatchDatset(self.valid_records, self.image_options)
def get_data(self):
with tf.device('/cpu:0'):
if self.mode == 'train':
self.train_dataset_reader = dataset.BatchDatset(self.train_records, self.image_options)
self.validation_dataset_reader = dataset.BatchDatset(self.valid_records, self.image_options)
for angle in range(0,360,60):
y_k.append(self.eval_net(
self.rotate( x[k*self.batch_size:(k+1)*self.batch_size],angle)
)
)
y_i.append(np.float32(y_k).mean(axis=0))
y = np.vstack(y_i)
return y
if __name__ == '__main__':
rng = numpy.random.RandomState(42)
br = BatchReader.inputs()
br2 = BatchReader.inputs(testingData = True)
X, Y = br.getNPArray(2)
testX = br2.getNPArray(2)
nkerns = [35,70,35]
print X.shape, Y.shape
print testX.shape
n_examples = X.shape[0]
fs = foldsize = n_examples/K_FOLD
# k folds
for k in range(K_FOLD):
trainX = np.vstack(
cur_frame = line[cfgs.seq_num + 1].split(' ')[0]
cur_frame_s = cur_frame.split("/")
filename = '%s%s' % (cur_frame_s[len(cur_frame_s) - 3],
os.path.splitext(cur_frame_s[-1])[0])
print(filename)
#anno_frame = os.path.join(anno_folder, filename+'.bmp')
record = {
'current': cur_frame,
'seq': seq_set,
'filename': filename
}
#print(record)
seq_list[directory].append(record)
random.shuffle(seq_list[directory])
no_of_images = len(seq_list[directory])
print('No of %s files %d' % (directory, no_of_images))
return seq_list
if __name__ == '__main__':
training_records, validation_records = my_read_dataset(
cfgs.seq_list_path, cfgs.anno_path)
image_options = {'resize': True, 'resize_size': 224}
train_reader = dataset.BatchDatset(training_records, image_options)
train_reader._read_images()
def main(argv=None):
keep_probability = tf.placeholder(tf.float32, name="keep_probabilty")
image = tf.placeholder(tf.float32,
shape=[None, IMAGE_SIZE, IMAGE_SIZE, 7],
name="input_image")
annotation = tf.placeholder(tf.int32,
shape=[None, IMAGE_SIZE, IMAGE_SIZE, 1],
name="annotation")
pred_annotation_value, pred_annotation, logits, pred_prob = inference(
image, keep_probability)
tf.summary.image("input_image", image, max_outputs=2)
tf.summary.image("ground_truth",
tf.cast(annotation, tf.uint8),
max_outputs=2)
tf.summary.image("pred_annotation",
tf.cast(pred_annotation, tf.uint8),
max_outputs=2)
#logits:the last layer of conv net
#labels:the ground truth
loss = tf.reduce_mean((tf.nn.sparse_softmax_cross_entropy_with_logits(
logits=logits,
labels=tf.squeeze(annotation, squeeze_dims=[3]),
name="entropy")))
tf.summary.scalar("entropy", loss)
trainable_var = tf.trainable_variables()
if cfgs.debug:
for var in trainable_var:
utils.add_to_regularization_and_summary(var)
print("Setting up summary op...")
summary_op = tf.summary.merge_all()
#Create a file to write logs.
#filename='logs'+ cfgs.mode + str(datetime.datetime.now()) + '.txt'
filename = "logs_%s%s.txt" % (cfgs.mode, datetime.datetime.now())
path_ = os.path.join(cfgs.logs_dir, filename)
logs_file = open(path_, 'w')
logs_file.write("The logs file is created at %s\n" %
datetime.datetime.now())
logs_file.write("The mode is %s\n" % (cfgs.mode))
logs_file.write(
"The train data batch size is %d and the validation batch size is %d.\n"
% (cfgs.batch_size, cfgs.v_batch_size))
logs_file.write("The train data is %s.\n" % (cfgs.data_dir))
logs_file.write("The model is ---%s---.\n" % cfgs.logs_dir)
print("Setting up image reader...")
logs_file.write("Setting up image reader...\n")
train_records, valid_records = scene_parsing.my_read_video_dataset(
cfgs.seq_list_path, cfgs.anno_path)
print('number of train_records', len(train_records))
print('number of valid_records', len(valid_records))
logs_file.write('number of train_records %d\n' % len(train_records))
logs_file.write('number of valid_records %d\n' % len(valid_records))
print("Setting up dataset reader")
vis = True if cfgs.mode == 'all_visualize' else False
image_options = {
'resize': True,
'resize_size': IMAGE_SIZE,
'visualize': vis
}
if cfgs.mode == 'train':
train_dataset_reader = dataset.BatchDatset(train_records,
image_options)
validation_dataset_reader = dataset.BatchDatset(valid_records,
image_options)
sess = tf.Session()
print("Setting up Saver...")
saver = tf.train.Saver()
summary_writer = tf.summary.FileWriter(cfgs.logs_dir, sess.graph)
sess.run(tf.global_variables_initializer())
ckpt = tf.train.get_checkpoint_state(cfgs.logs_dir)
#if not train,restore the model trained before
if ckpt and ckpt.model_checkpoint_path:
saver.restore(sess, ckpt.model_checkpoint_path)
print("Model restored...")
if cfgs.mode == "accurary":
count = 0
if_con = True
accu_iou_t = 0
accu_pixel_t = 0
while if_con:
count = count + 1
valid_images, valid_annotations, valid_filenames, if_con, start, end = validation_dataset_reader.next_batch_valid(
cfgs.v_batch_size)
valid_loss, pred_anno = sess.run(
[loss, pred_annotation],
feed_dict={
image: valid_images,
annotation: valid_annotations,
keep_probability: 1.0
})
accu_iou, accu_pixel = accu.caculate_accurary(
pred_anno, valid_annotations)
print("Ture %d ---> the data from %d to %d" % (count, start, end))
print("%s ---> Validation_pixel_accuary: %g" %
(datetime.datetime.now(), accu_pixel))
print("%s ---> Validation_iou_accuary: %g" %
(datetime.datetime.now(), accu_iou))
#Output logs.
logs_file.write("Ture %d ---> the data from %d to %d\n" %
(count, start, end))
logs_file.write("%s ---> Validation_pixel_accuary: %g\n" %
(datetime.datetime.now(), accu_pixel))
logs_file.write("%s ---> Validation_iou_accuary: %g\n" %
(datetime.datetime.now(), accu_iou))
accu_iou_t = accu_iou_t + accu_iou
accu_pixel_t = accu_pixel_t + accu_pixel
print("%s ---> Total validation_pixel_accuary: %g" %
(datetime.datetime.now(), accu_pixel_t / count))
print("%s ---> Total validation_iou_accuary: %g" %
(datetime.datetime.now(), accu_iou_t / count))
#Output logs
logs_file.write("%s ---> Total validation_pixel_accurary: %g\n" %
(datetime.datetime.now(), accu_pixel_t / count))
logs_file.write("%s ---> Total validation_iou_accurary: %g\n" %
(datetime.datetime.now(), accu_iou_t / count))
elif cfgs.mode == "all_visualize":
re_save_dir = "%s%s" % (cfgs.result_dir, datetime.datetime.now())
logs_file.write("The result is save at file'%s'.\n" % re_save_dir)
logs_file.write("The number of part visualization is %d.\n" %
cfgs.v_batch_size)
#Check the result path if exists.
if not os.path.exists(re_save_dir):
print("The path '%s' is not found." % re_save_dir)
print("Create now ...")
os.makedirs(re_save_dir)
print("Create '%s' successfully." % re_save_dir)
logs_file.write("Create '%s' successfully.\n" % re_save_dir)
re_save_dir_im = os.path.join(re_save_dir, 'images')
re_save_dir_heat = os.path.join(re_save_dir, 'heatmap')
re_save_dir_ellip = os.path.join(re_save_dir, 'ellip')
re_save_dir_transheat = os.path.join(re_save_dir, 'transheat')
if not os.path.exists(re_save_dir_im):
os.makedirs(re_save_dir_im)
if not os.path.exists(re_save_dir_heat):
os.makedirs(re_save_dir_heat)
if not os.path.exists(re_save_dir_ellip):
os.makedirs(re_save_dir_ellip)
if not os.path.exists(re_save_dir_transheat):
os.makedirs(re_save_dir_transheat)
count = 0
if_con = True
accu_iou_t = 0
accu_pixel_t = 0
while if_con:
count = count + 1
valid_images, valid_filename, valid_cur_images, if_con, start, end = validation_dataset_reader.next_batch_video_valid(
cfgs.v_batch_size)
pred_value, pred, logits_, pred_prob_ = sess.run(
[pred_annotation_value, pred_annotation, logits, pred_prob],
feed_dict={
image: valid_images,
keep_probability: 1.0
})
print("Turn %d :----start from %d ------- to %d" %
(count, start, end))
pred = np.squeeze(pred, axis=3)
pred_value = np.squeeze(pred_value, axis=3)
for itr in range(len(pred)):
filename = valid_filename[itr]['filename']
valid_images_ = pred_visualize(valid_cur_images[itr].copy(),
pred[itr])
utils.save_image(valid_images_.astype(np.uint8),
re_save_dir_im,
name="inp_" + filename)
if cfgs.fit_ellip:
#valid_images_ellip=fit_ellipse_findContours_ori(valid_images[itr].copy(),np.expand_dims(pred[itr],axis=2).astype(np.uint8))
valid_images_ellip = fit_ellipse_findContours(
valid_cur_images[itr].copy(),
np.expand_dims(pred[itr], axis=2).astype(np.uint8))
utils.save_image(valid_images_ellip.astype(np.uint8),
re_save_dir_ellip,
name="ellip_" + filename)
if cfgs.heatmap:
heat_map = density_heatmap(pred_prob_[itr, :, :, 1])
utils.save_image(heat_map.astype(np.uint8),
re_save_dir_heat,
name="heat_" + filename)
if cfgs.trans_heat:
trans_heat_map = translucent_heatmap(
valid_cur_images[itr],
heat_map.astype(np.uint8).copy())
utils.save_image(trans_heat_map,
re_save_dir_transheat,
name="trans_heat_" + filename)
import BatchReader as dataset
import os
import time
from CaptchaModel import Captcha
# load model
model = Captcha()
model.load_checkpoint('crack_capcha0.994400002956.model-9960')
# load dataset
dataset = dataset.BatchReader(ratio=0.01, test_size=1000)
images, labels = dataset.get_val_batch(0, 1000)
count = len(labels)
correct = 0
# start timing
start = time.time()
t = start
# start testing
for i in range(count):
image = images[i]
label = labels[i]
pred = model.predict(image)
text = ''.join(map(str, pred))
if text == label:
correct += 1
else:
print(label, text)
pass
for var in trainable_var:
utils.add_to_regularization_and_summary(var)
train_op = train(loss, trainable_var)
print("Setting up summary op...")
summary_op = tf.summary.merge_all()
print("Setting up image reader...")
train_records, valid_records = scene_parsing.read_dataset(FLAGS.data_dir)
print(len(train_records))
print(len(valid_records))
print("Setting up dataset reader")
image_options = {'resize': True, 'resize_size': IMAGE_SIZE}
if FLAGS.mode == 'train':
train_dataset_reader = dataset.BatchDatset(train_records, image_options)
validation_dataset_reader = dataset.BatchDatset(valid_records, image_options)
sess = tf.Session()
print("Setting up Saver...")
saver = tf.train.Saver( max_to_keep=10)
summary_writer = tf.summary.FileWriter(FLAGS.logs_dir+'/train', sess.graph)
# tval_summary_writer = tf.summary.FileWriter(FLAGS.logs_dir + '/val', sess.graph)
sess.run(tf.global_variables_initializer())
ckpt = tf.train.get_checkpoint_state(FLAGS.logs_dir)
if ckpt and ckpt.model_checkpoint_path:
saver.restore(sess, ckpt.model_checkpoint_path)
print("Model restored...")
print(__doc__)
from sklearn import neighbors, linear_model
import BatchReader
reader = BatchReader.inputs()
array = reader.getNPArray(7809)
X_digits = array[0]
y_digits = array[1]
n_samples = len(X_digits)
X_train = X_digits[:.9*n_samples]
y_train = y_digits[:.9*n_samples]
X_test = X_digits[.9*n_samples:]
y_test = y_digits[.9*n_samples:]
logistic = linear_model.LogisticRegression()
print('LogisticRegression score: %f' % logistic.fit(X_train, y_train).score(X_test,y_test))
def evaluate_lenet5(learning_rate=0.1,
n_epochs=200,
dataset='mnist.pkl.gz',
nkerns=[20, 50],
batch_size=500):
""" Demonstrates lenet on MNIST dataset
:type learning_rate: float
:param learning_rate: learning rate used (factor for the stochastic
gradient)
:type n_epochs: int
:param n_epochs: maximal number of epochs to run the optimizer
:type dataset: string
:param dataset: path to the dataset used for training /testing (MNIST here)
:type nkerns: list of ints
:param nkerns: number of kernels on each layer
"""
rng = numpy.random.RandomState(23455)
br = BatchReader.inputs()
br2 = BatchReader.inputs(testingData=True)
X, Y = br.getNPArray(2)
train_set_x = theano.shared(X[0:50000 * 0.6].astype(float))
train_set_y = theano.shared(Y[0:50000 * 0.6].astype('int32'))
valid_set_x = theano.shared(X[50000 * 0.6:50000 * 0.9].astype(float))
valid_set_y = theano.shared(Y[50000 * 0.6:50000 * 0.9].astype('int32'))
test_set_x = theano.shared(X[50000 * 0.9:50000].astype(float))
test_set_y = theano.shared(Y[50000 * 0.9:50000].astype('int32'))
#datasets = load_data(dataset)
#train_set_x, train_set_y = datasets[0]
#valid_set_x, valid_set_y = datasets[1]
#test_set_x, test_set_y = datasets[2]
# compute number of minibatches for training, validation and testing
n_train_batches = train_set_x.get_value(borrow=True).shape[0]
n_valid_batches = valid_set_x.get_value(borrow=True).shape[0]
n_test_batches = test_set_x.get_value(borrow=True).shape[0]
n_train_batches /= batch_size
n_valid_batches /= batch_size
n_test_batches /= batch_size
# allocate symbolic variables for the data
index = T.lscalar() # index to a [mini]batch
# start-snippet-1
x = T.matrix('x') # the data is presented as rasterized images
y = T.ivector('y') # the labels are presented as 1D vector of
# [int] labels
######################
# BUILD ACTUAL MODEL #
######################
print '... building the model'
# Reshape matrix of rasterized images of shape (batch_size, 28 * 28)
# to a 4D tensor, compatible with our LeNetConvPoolLayer
# (28, 28) is the size of MNIST images.
layer0_input = x.reshape((batch_size, 1, 48, 48))
# Construct the first convolutional pooling layer:
# filtering reduces the image size to (28-5+1 , 28-5+1) = (24, 24)
# maxpooling reduces this further to (24/2, 24/2) = (12, 12)
# 4D output tensor is thus of shape (batch_size, nkerns[0], 12, 12)
layer0 = LeNetConvPoolLayer(rng,
input=layer0_input,
image_shape=(batch_size, 1, 48, 48),
filter_shape=(nkerns[0], 1, 9, 9),
poolsize=(2, 2))
# Construct the second convolutional pooling layer
# filtering reduces the image size to (12-5+1, 12-5+1) = (8, 8)
# maxpooling reduces this further to (8/2, 8/2) = (4, 4)
# 4D output tensor is thus of shape (batch_size, nkerns[1], 4, 4)
layer1 = LeNetConvPoolLayer(rng,
input=layer0.output,
image_shape=(batch_size, nkerns[0], 20, 20),
filter_shape=(nkerns[1], nkerns[0], 8, 8),
poolsize=(2, 2))
# the HiddenLayer being fully-connected, it operates on 2D matrices of
# shape (batch_size, num_pixels) (i.e matrix of rasterized images).
# This will generate a matrix of shape (batch_size, nkerns[1] * 4 * 4),
# or (500, 50 * 4 * 4) = (500, 800) with the default values.
layer2_input = layer1.output.flatten(2)
# construct a fully-connected sigmoidal layer
layer2 = HiddenLayer(rng,
input=layer2_input,
n_in=nkerns[1] * 6 * 6,
n_out=500,
activation=T.tanh)
# classify the values of the fully-connected sigmoidal layer
layer3 = LogisticRegression(input=layer2.output, n_in=500, n_out=10)
# the cost we minimize during training is the NLL of the model
cost = layer3.negative_log_likelihood(y)
print type(test_set_x)
# create a function to compute the mistakes that are made by the model
test_model = theano.function(
[index],
layer3.errors(y),
givens={
x: test_set_x[index * batch_size:(index + 1) * batch_size],
y: test_set_y[index * batch_size:(index + 1) * batch_size]
})
validate_model = theano.function(
[index],
layer3.errors(y),
givens={
x: valid_set_x[index * batch_size:(index + 1) * batch_size],
y: valid_set_y[index * batch_size:(index + 1) * batch_size]
})
# create a list of all model parameters to be fit by gradient descent
params = layer3.params + layer2.params + layer1.params + layer0.params
# create a list of gradients for all model parameters
grads = T.grad(cost, params)
# train_model is a function that updates the model parameters by
# SGD Since this model has many parameters, it would be tedious to
# manually create an update rule for each model parameter. We thus
# create the updates list by automatically looping over all
# (params[i], grads[i]) pairs.
updates = [(param_i, param_i - learning_rate * grad_i)
for param_i, grad_i in zip(params, grads)]
train_model = theano.function(
[index],
cost,
updates=updates,
givens={
x: train_set_x[index * batch_size:(index + 1) * batch_size],
y: train_set_y[index * batch_size:(index + 1) * batch_size]
})
# end-snippet-1
###############
# TRAIN MODEL #
###############
print '... training'
# early-stopping parameters
patience = 10000 # look as this many examples regardless
patience_increase = 2 # wait this much longer when a new best is
# found
improvement_threshold = 0.995 # a relative improvement of this much is
# considered significant
validation_frequency = min(n_train_batches, patience / 2)
# go through this many
# minibatche before checking the network
# on the validation set; in this case we
# check every epoch
best_validation_loss = numpy.inf
best_iter = 0
test_score = 0.
start_time = timeit.default_timer()
epoch = 0
done_looping = False
while (epoch < n_epochs) and (not done_looping):
epoch = epoch + 1
for minibatch_index in xrange(n_train_batches):
iter = (epoch - 1) * n_train_batches + minibatch_index
if iter % 100 == 0:
print 'training @ iter = ', iter
cost_ij = train_model(minibatch_index)
if (iter + 1) % validation_frequency == 0:
# compute zero-one loss on validation set
validation_losses = [
validate_model(i) for i in xrange(n_valid_batches)
]
this_validation_loss = numpy.mean(validation_losses)
print('epoch %i, minibatch %i/%i, validation error %f %%' %
(epoch, minibatch_index + 1, n_train_batches,
this_validation_loss * 100.))
# if we got the best validation score until now
if this_validation_loss < best_validation_loss:
#improve patience if loss improvement is good enough
if this_validation_loss < best_validation_loss * \
improvement_threshold:
patience = max(patience, iter * patience_increase)
# save best validation score and iteration number
best_validation_loss = this_validation_loss
best_iter = iter
# test it on the test set
test_losses = [
test_model(i) for i in xrange(n_test_batches)
]
test_score = numpy.mean(test_losses)
print((' epoch %i, minibatch %i/%i, test error of '
'best model %f %%') %
(epoch, minibatch_index + 1, n_train_batches,
test_score * 100.))
if patience <= iter:
done_looping = True
break
end_time = timeit.default_timer()
print('Optimization complete.')
print(
'Best validation score of %f %% obtained at iteration %i, '
'with test performance %f %%' %
(best_validation_loss * 100., best_iter + 1, test_score * 100.))
print >> sys.stderr, ('The code for file ' + os.path.split(__file__)[1] +
' ran for %.2fm' % ((end_time - start_time) / 60.))
print "Dumping parameters to ../data/convnet.pkl"
dir = os.getcwd()
path = os.path.join(dir, "../data")
os.chdir(path)
with open("convnet.pkl", 'wb') as file:
dict = {
"layer0": layer0.param,
"layer1": layer1.param,
"layer2": layer2.param,
"layer3": layer3.param
}
pickle.dump(dict, file, 2)
os.chdir(dir)
def get_reader(filename):
arc = ZipTarReader.auto(filename)
return BatchReader.BatchReader(arc, arc.namelist(), KaggleLabelClassifier)
y_k = []
for angle in range(0, 360, 60):
y_k.append(
self.eval_net(
self.rotate(
x[k * self.batch_size:(k + 1) * self.batch_size],
angle)))
y_i.append(np.float32(y_k).mean(axis=0))
y = np.vstack(y_i)
return y
if __name__ == '__main__':
rng = numpy.random.RandomState(42)
br = BatchReader.inputs()
br2 = BatchReader.inputs(testingData=True)
X, Y = br.getNPArray(2)
testX = br2.getNPArray(2)
nkerns = [35, 70, 35]
print X.shape, Y.shape
print testX.shape
n_examples = X.shape[0]
fs = foldsize = n_examples / K_FOLD
# k folds
for k in range(K_FOLD):
trainX = np.vstack([X[:k * fs], X[(k + 1) * fs:]])
