0
我是張量流的新人，我正在訓練移動net_v1。要做到這一點，我首先從一個txt文件中創建的tfrecords'多類文件（例如：namefile LABEL1 LABEL2 ...）來自logits和檢查點的訓練模型
import sys, os
import tensorflow as tf
import cv2
import numpy as np
import matplotlib.pyplot as plt
# function
def load_image(addr):
# read an image and resize to (224, 224)
# cv2 load images as BGR, convert it to RGB
img = cv2.imread(addr)
img = cv2.resize(img, (224, 224), interpolation=cv2.INTER_CUBIC)
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
img = img.astype(np.float32)
return img
def _int64_feature(value):
return tf.train.Feature(int64_list=tf.train.Int64List(value=[*value]))
def _bytes_feature(value):
return tf.train.Feature(bytes_list=tf.train.BytesList(value=[value]))
def loadData(inputs):
addrs = []
labels = []
f = open(inputs, 'r')
data = [ln.split(' ') for ln in f ]
f.close()
print(data)
for i in range(0, len(data)):
addrs.append(data[i][0].rstrip())
l = []
for j in range(1,len(data[i])):
if(data[i][j].rstrip().isdigit() == True):
l.append(int(data[i][j].rstrip()))
print(l)
labels.append(l)
return addrs, labels
def CreateTrainFile(input_filename, train_filename,):
path = '/home/rd/Documents/RD2/Databases/Faces/'
# load file and label
train_addrs, train_labels = loadData(input_filename)
print(train_labels)
# open the TFRecords file
writer = tf.python_io.TFRecordWriter(train_filename)
for i in range(len(train_addrs)):
# print how many images are saved every 1000 images
if not i % 1000:
print('Train data: {}/{}'.format(i, len(train_addrs)))
sys.stdout.flush()
# Load the image
img = load_image(train_addrs[i])
label = train_labels[i]
print('label : ', _int64_feature(label))
# Create a feature
feature = {'train/label': _int64_feature(label),
'train/image': _bytes_feature(tf.compat.as_bytes(img.tostring()))}
# Create an example protocol buffer
example = tf.train.Example(features=tf.train.Features(feature=feature))
# Serialize to string and write on the file
writer.write(example.SerializeToString())
writer.close()
sys.stdout.flush()
# open the TFRecords file
def CreateValidationFile(val_filename):
writer = tf.python_io.TFRecordWriter(val_filename)
for i in range(len(val_addrs)):
# print how many images are saved every 1000 images
if not i % 1000:
print('Val data: {}/{}'.format(i, len(val_addrs)))
sys.stdout.flush()
# Load the image
img = load_image(val_addrs[i])
label = val_labels[i]
# Create a feature
feature = {'val/label': _int64_feature(label),
'val/image': _bytes_feature(tf.compat.as_bytes(img.tostring()))}
# Create an example protocol buffer
example = tf.train.Example(features=tf.train.Features(feature=feature))
# Serialize to string and write on the file
writer.write(example.SerializeToString())
writer.close()
sys.stdout.flush()
# open the TFRecords file
def CreateTestFile(test_filename):
writer = tf.python_io.TFRecordWriter(test_filename)
for i in range(len(test_addrs)):
# print how many images are saved every 1000 images
if not i % 1000:
print('Test data: {}/{}'.format(i, len(test_addrs)))
sys.stdout.flush()
# Load the image
img = load_image(test_addrs[i])
label = test_labels[i]
# Create a feature
feature = {'test/label': _int64_feature(label),
'test/image': _bytes_feature(tf.compat.as_bytes(img.tostring()))}
# Create an example protocol buffer
example = tf.train.Example(features=tf.train.Features(feature=feature))
# Serialize to string and write on the file
writer.write(example.SerializeToString())
writer.close()
sys.stdout.flush()
def ReadRecordFileTrain(data_path):
#data_path = 'train.tfrecords' # address to save the hdf5 file
with tf.Session() as sess:
feature = {'train/image': tf.FixedLenFeature([], tf.string),
'train/label': tf.FixedLenFeature([], tf.int64)}
# Create a list of filenames and pass it to a queue
filename_queue = tf.train.string_input_producer([data_path], num_epochs=1)
# Define a reader and read the next record
reader = tf.TFRecordReader()
_, serialized_example = reader.read(filename_queue)
# Decode the record read by the reader
features = tf.parse_single_example(serialized_example, features=feature)
# Convert the image data from string back to the numbers
image = tf.decode_raw(features['train/image'], tf.float32)
# Cast label data into int32
label = tf.cast(features['train/label'], tf.int32)
# Reshape image data into the original shape
image = tf.reshape(image, [224, 224, 3])
# Any preprocessing here ...
# Creates batches by randomly shuffling tensors
images, labels = tf.train.shuffle_batch([image, label], batch_size=2, capacity=30, num_threads=1, min_after_dequeue=10)
return images, labels
def main():
train_filename = 'train.tfrecords' # address to save the TFRecords file
#test_filename = 'test.tfrecords' # address to save the TFRecords file
#val_filename = 'val.tfrecords' # address to save the TFRecords file
CreateTrainFile("data.txt", train_filename)
main()
和讀取TF記錄：
def ReadRecordFileTrain(data_path):
#data_path = 'train.tfrecords' # address to save the hdf5 file
with tf.Session() as sess:
feature = {'train/image': tf.FixedLenFeature([], tf.string),
'train/label': tf.FixedLenFeature([2], tf.int64)}
# Create a list of filenames and pass it to a queue
filename_queue = tf.train.string_input_producer([data_path], num_epochs=1)
# Define a reader and read the next record
reader = tf.TFRecordReader()
_, serialized_example = reader.read(filename_queue)
# Decode the record read by the reader
features = tf.parse_single_example(serialized_example, features=feature)
# Convert the image data from string back to the numbers
image = tf.decode_raw(features['train/image'], tf.float32)
print('label1 :', features['train/label'])
# Cast label data into int32
label = tf.cast(features['train/label'], tf.int32)
print('label load:', label)
# Reshape image data into the original shape
image = tf.reshape(image, [224, 224, 3])
# Any preprocessing here ...
# Creates batches by randomly shuffling tensors
images, labels = tf.train.batch([image, label], batch_size=2, capacity=30, num_threads=1)
return images, labels
我想它可以工作，但我不確定（當我調用這些函數時，我沒有任何錯誤）。然後，我加載模型及其重量。調用損失函數並嘗試開始訓練，但此時失敗。
g = tf.Graph()
with g.as_default():
# size of the folder
inputs = tf.placeholder(tf.float32, [1, 224, 224, 3])
# load dataset
images, labels = ReadRecordFileTrain('train.tfrecords')
print('load dataset done')
print('labels = ', labels)
print('data = ', images)
print(tf.shape(labels))
# load network
network, end_points= mobilenet.mobilenet_v1(images, num_classes=2, depth_multiplier=0.25)
print('load network done')
print('network : ', network)
variables_to_restore = slim.get_variables_to_restore(exclude=["MobilenetV1/Logits/Conv2d_1c_1x1"])
load_checkpoint = "modele_mobilenet_v1_025/mobilenet_v1_0.25_224.ckpt"
init_fn = slim.assign_from_checkpoint_fn(load_checkpoint, variables_to_restore)
print('custom network done')
# Specify the loss function:
tf.losses.softmax_cross_entropy(labels, network)
total_loss = tf.losses.get_total_loss()
#tf.scalar_summary('losses/total_loss', total_loss)
# Specify the optimization scheme:
optimizer = tf.train.GradientDescentOptimizer(learning_rate=.001)
# create_train_op that ensures that when we evaluate it to get the loss,
# the update_ops are done and the gradient updates are computed.
train_tensor = slim.learning.create_train_op(total_loss, optimizer)
print('loss and optimizer chosen')
# Actually runs training.
save_checkpoint = 'model/modelcheck'
# start training
learning = slim.learning.train(train_tensor, save_checkpoint, init_fn=init_fn, number_of_steps=1000)
錯誤消息：
label1 : Tensor("ParseSingleExample/Squeeze_train/label:0", shape=(2,), dtype=int64)
label load: Tensor("Cast:0", shape=(2,), dtype=int32)
load dataset done
labels = Tensor("batch:1", shape=(2, 2), dtype=int32)
data = Tensor("batch:0", shape=(2, 224, 224, 3), dtype=float32)
Tensor("Shape:0", shape=(2,), dtype=int32)
load network done
network : Tensor("MobilenetV1/Logits/SpatialSqueeze:0", shape=(2, 2), dtype=float32)
custom network done
loss and optimizer chosen
Traceback (most recent call last):
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 1039, in _do_call
return fn(*args)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 1021, in _run_fn
status, run_metadata)
File "/usr/lib/python3.5/contextlib.py", line 66, in __exit__
next(self.gen)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/errors_impl.py", line 466, in raise_exception_on_not_ok_status
pywrap_tensorflow.TF_GetCode(status))
tensorflow.python.framework.errors_impl.InvalidArgumentError: Assign requires shapes of both tensors to match. lhs shape= [1,1,256,2] rhs shape= [1,1,256,1]
[[Node: save_1/Assign_109 = Assign[T=DT_FLOAT, _class=["loc:@MobilenetV1/Logits/Conv2d_1c_1x1/weights"], use_locking=true, validate_shape=true, _device="/job:localhost/replica:0/task:0/cpu:0"](MobilenetV1/Logits/Conv2d_1c_1x1/weights, save_1/RestoreV2_109)]]
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "test.py", line 106, in <module>
main()
File "test.py", line 103, in main
learning = slim.learning.train(train_tensor, save_checkpoint, init_fn=init_fn, number_of_steps=1000)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/contrib/slim/python/slim/learning.py", line 725, in train
master, start_standard_services=False, config=session_config) as sess:
File "/usr/lib/python3.5/contextlib.py", line 59, in __enter__
return next(self.gen)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/supervisor.py", line 960, in managed_session
self.stop(close_summary_writer=close_summary_writer)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/supervisor.py", line 788, in stop
stop_grace_period_secs=self._stop_grace_secs)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/coordinator.py", line 389, in join
six.reraise(*self._exc_info_to_raise)
File "/usr/lib/python3/dist-packages/six.py", line 686, in reraise
raise value
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/supervisor.py", line 949, in managed_session
start_standard_services=start_standard_services)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/supervisor.py", line 706, in prepare_or_wait_for_session
init_feed_dict=self._init_feed_dict, init_fn=self._init_fn)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/session_manager.py", line 256, in prepare_session
config=config)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/session_manager.py", line 188, in _restore_checkpoint
saver.restore(sess, ckpt.model_checkpoint_path)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 1457, in restore
{self.saver_def.filename_tensor_name: save_path})
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 778, in run
run_metadata_ptr)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 982, in _run
feed_dict_string, options, run_metadata)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 1032, in _do_run
target_list, options, run_metadata)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/client/session.py", line 1052, in _do_call
raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InvalidArgumentError: Assign requires shapes of both tensors to match. lhs shape= [1,1,256,2] rhs shape= [1,1,256,1]
[[Node: save_1/Assign_109 = Assign[T=DT_FLOAT, _class=["loc:@MobilenetV1/Logits/Conv2d_1c_1x1/weights"], use_locking=true, validate_shape=true, _device="/job:localhost/replica:0/task:0/cpu:0"](MobilenetV1/Logits/Conv2d_1c_1x1/weights, save_1/RestoreV2_109)]]
Caused by op 'save_1/Assign_109', defined at:
File "test.py", line 106, in <module>
main()
File "test.py", line 103, in main
learning = slim.learning.train(train_tensor, save_checkpoint, init_fn=init_fn, number_of_steps=1000)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/contrib/slim/python/slim/learning.py", line 642, in train
saver = saver or tf_saver.Saver()
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 1056, in __init__
self.build()
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 1086, in build
restore_sequentially=self._restore_sequentially)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 691, in build
restore_sequentially, reshape)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 419, in _AddRestoreOps
assign_ops.append(saveable.restore(tensors, shapes))
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/training/saver.py", line 155, in restore
self.op.get_shape().is_fully_defined())
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/state_ops.py", line 270, in assign
validate_shape=validate_shape)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/gen_state_ops.py", line 47, in assign
use_locking=use_locking, name=name)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/op_def_library.py", line 768, in apply_op
op_def=op_def)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py", line 2336, in create_op
original_op=self._default_original_op, op_def=op_def)
File "/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py", line 1228, in __init__
self._traceback = _extract_stack()
InvalidArgumentError (see above for traceback): Assign requires shapes of both tensors to match. lhs shape= [1,1,256,2] rhs shape= [1,1,256,1]
[[Node: save_1/Assign_109 = Assign[T=DT_FLOAT, _class=["loc:@MobilenetV1/Logits/Conv2d_1c_1x1/weights"], use_locking=true, validate_shape=true, _device="/job:localhost/replica:0/task:0/cpu:0"](MobilenetV1/Logits/Conv2d_1c_1x1/weights, save_1/RestoreV2_109)]]
我不明白問題出在哪裏的來源和如何解決它。
我曾經得到這個錯誤，當我在模型中保存目錄模型與我的當前運行模式的衝突。嘗試刪除模型目錄並重新開始訓練。 –
謝謝，它解決了這個問題！ – Shiro