我已经使用mnist训练数据和所有这些来复制Tensorflow教程中基于逐字逐句使用tf.layers的代码。培训和评估代码,预测字典和评估指标都保持不变。我遇到的问题是,我已经在我自己的所有图像上获得了非常不规则的输出,我试图将其输入网络并获得预测。到目前为止,我遇到了两个主要问题:
1)我单独提供给网络的每个图像(在修改它们之后,如第2点所示)不断给我一个输出:class [8]。这意味着网络将其识别为数字' 8'不是吗?我尝试输入4,7,0等等,但每次读取为8时。奇怪的是,经过培训,网络声称具有非常高的准确度等级,但它没有甚至为不同的输入数字提供不同的输出。我想知道我到底出了什么问题。
2)我正在使用opencv imread函数读取.jpg图像并首先将其转换为float 64(使用scikit-image),然后将我的图像转换为具有numpy astype的float 32图像。我这样做是为了让预测真正得以实现。出于某种原因,网络拒绝预测我的常规uint8图像,该图像最初由opencv读取。错误消息表明它只会将float16,bfloat16或float32类型的图像作为输入。
无论如何,这就是为什么我试图尽可能地调整原始图像,并在视觉上验证图像与原始输入图像相比似乎没有变化。因此,使用float32图像输入,程序给了我输出,但它们都是相同的(都是类[8]),我认为我在代码中的某个地方犯了一个错误。
这是源代码,我自己的tf.estimator.predict()。我在Windows 10上使用的是Python 3.6.3,如果相关则为64位。
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import numpy as np
import tensorflow as tf
import cv2
import skimage
tf.logging.set_verbosity(tf.logging.INFO)
image1=cv2.imread("E:\Predict_images\image (5).jpg")
image2=skimage.util.img_as_float(image1)
image3=image2.astype(np.float32)
cv2.imshow('image1',image1)
cv2.imshow('image2',image2)
cv2.imshow('image 3',image3)
def cnn_model_fn(features, labels, mode):
"""Model function for CNN."""
# Input Layer
# Reshape X to 4-D tensor: [batch_size, width, height, channels]
# MNIST images are 28x28 pixels, and have one color channel
input_layer = tf.reshape(features["x"], [-1, 28, 28, 1])
# Convolutional Layer #1
# Computes 32 features using a 5x5 filter with ReLU activation.
# Padding is added to preserve width and height.
# Input Tensor Shape: [batch_size, 28, 28, 1]
# Output Tensor Shape: [batch_size, 28, 28, 32]
conv1 = tf.layers.conv2d(
inputs=input_layer,
filters=32,
kernel_size=[5, 5],
padding="same",
activation=tf.nn.relu)
# Pooling Layer #1
# First max pooling layer with a 2x2 filter and stride of 2
# Input Tensor Shape: [batch_size, 28, 28, 32]
# Output Tensor Shape: [batch_size, 14, 14, 32]
pool1 = tf.layers.max_pooling2d(inputs=conv1, pool_size=[2, 2], strides=2)
# Convolutional Layer #2
# Computes 64 features using a 5x5 filter.
# Padding is added to preserve width and height.
# Input Tensor Shape: [batch_size, 14, 14, 32]
# Output Tensor Shape: [batch_size, 14, 14, 64]
conv2 = tf.layers.conv2d(
inputs=pool1,
filters=64,
kernel_size=[5, 5],
padding="same",
activation=tf.nn.relu)
# Pooling Layer #2
# Second max pooling layer with a 2x2 filter and stride of 2
# Input Tensor Shape: [batch_size, 14, 14, 64]
# Output Tensor Shape: [batch_size, 7, 7, 64]
pool2 = tf.layers.max_pooling2d(inputs=conv2, pool_size=[2, 2], strides=2)
# Flatten tensor into a batch of vectors
# Input Tensor Shape: [batch_size, 7, 7, 64]
# Output Tensor Shape: [batch_size, 7 * 7 * 64]
pool2_flat = tf.reshape(pool2, [-1, 7 * 7 * 64])
# Dense Layer
# Densely connected layer with 1024 neurons
# Input Tensor Shape: [batch_size, 7 * 7 * 64]
# Output Tensor Shape: [batch_size, 1024]
dense = tf.layers.dense(inputs=pool2_flat, units=1024, activation=tf.nn.relu)
# Add dropout operation; 0.6 probability that element will be kept
dropout = tf.layers.dropout(
inputs=dense, rate=0.4, training=mode == tf.estimator.ModeKeys.TRAIN)
# Logits layer
# Input Tensor Shape: [batch_size, 1024]
# Output Tensor Shape: [batch_size, 10]
logits = tf.layers.dense(inputs=dropout, units=10)
predictions = {
# Generate predictions (for PREDICT and EVAL mode)
"classes": tf.argmax(input=logits, axis=1),
# Add `softmax_tensor` to the graph. It is used for PREDICT and by the
# `logging_hook`.
"probabilities": tf.nn.softmax(logits, name="softmax_tensor")
}
if mode == tf.estimator.ModeKeys.PREDICT:
return tf.estimator.EstimatorSpec(mode=mode, predictions=predictions)
# Calculate Loss (for both TRAIN and EVAL modes)
loss = tf.losses.sparse_softmax_cross_entropy(labels=labels, logits=logits)
# Configure the Training Op (for TRAIN mode)
if mode == tf.estimator.ModeKeys.TRAIN:
optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.001)
train_op = optimizer.minimize(
loss=loss,
global_step=tf.train.get_global_step())
return tf.estimator.EstimatorSpec(mode=mode, loss=loss, train_op=train_op)
# Add evaluation metrics (for EVAL mode)
eval_metric_ops = {
"accuracy": tf.metrics.accuracy(
labels=labels, predictions=predictions["classes"])}
return tf.estimator.EstimatorSpec(
mode=mode, loss=loss, eval_metric_ops=eval_metric_ops)
def main(unused_argv):
# Load training and eval data
mnist = tf.contrib.learn.datasets.load_dataset("mnist")
train_data = mnist.train.images # Returns np.array
train_labels = np.asarray(mnist.train.labels, dtype=np.int32)
eval_data = mnist.test.images # Returns np.array
eval_labels = np.asarray(mnist.test.labels, dtype=np.int32)
# Create the Estimator
mnist_classifier = tf.estimator.Estimator(
model_fn=cnn_model_fn, model_dir="/tmp/mnist_convnet_model")
# Set up logging for predictions
# Log the values in the "Softmax" tensor with label "probabilities"
tensors_to_log = {"probabilities": "softmax_tensor"}
logging_hook = tf.train.LoggingTensorHook(
tensors=tensors_to_log, every_n_iter=50)
# Train the model
train_input_fn = tf.estimator.inputs.numpy_input_fn(
x={"x": train_data},
y=train_labels,
batch_size=100,
num_epochs=None,
shuffle=True)
mnist_classifier.train(
input_fn=train_input_fn,
steps=20000,
hooks=[logging_hook])
# Evaluate the model and print results
eval_input_fn = tf.estimator.inputs.numpy_input_fn(
x={"x": eval_data},
y=eval_labels,
num_epochs=1,
shuffle=False)
eval_results = mnist_classifier.evaluate(input_fn=eval_input_fn)
print(eval_results)
#Predict the result for new images
pred_input_fn=tf.estimator.inputs.numpy_input_fn(
x={"x": image3},
shuffle=False)
pred = mnist_classifier.predict(input_fn=pred_input_fn)
print (list(pred))
if __name__ == "__main__":
tf.app.run()
当我通过它运行这个It's an image of '1' float 32图像时,这是shell中给出的输出:
INFO:tensorflow:Saving checkpoints for 20000 into /tmp/mnist_convnet_model\model.ckpt.
INFO:tensorflow:Loss for final step: 0.040543813.
INFO:tensorflow:Starting evaluation at 2018-02-25-19:40:57
INFO:tensorflow:Restoring parameters from /tmp/mnist_convnet_model\model.ckpt-20000
INFO:tensorflow:Finished evaluation at 2018-02-25-19:40:59
INFO:tensorflow:Saving dict for global step 20000: accuracy = 0.9717, global_step = 20000, loss = 0.09776818
{'accuracy': 0.9717, 'loss': 0.09776818, 'global_step': 20000}
INFO:tensorflow:Restoring parameters from /tmp/mnist_convnet_model\model.ckpt-20000
[{'classes': 8, 'probabilities': array([4.6938831e-07, 3.4720117e-07, 5.8355248e-03, 7.6848278e-03,
1.0895459e-06, 2.9385969e-06, 1.8598693e-06, 1.2013125e-09,
9.8647296e-01, 5.2439987e-08], dtype=float32)}, {'classes': 8, 'probabilities': array([4.8870197e-07, 3.7765903e-07, 6.4324187e-03, 8.6945957e-03,
6.3728720e-07, 3.2801559e-06, 1.5783016e-06, 1.3214099e-09,
9.8486656e-01, 5.1528065e-08], dtype=float32)}, {'classes': 8, 'probabilities': array([5.9954516e-07, 3.2683863e-07, 7.1799601e-03, 8.4864357e-03,
8.9206560e-07, 2.3296870e-06, 1.4247992e-06, 9.8779152e-10,
9.8432791e-01, 7.0583980e-08], dtype=float32)}]
编辑:我只是留下了一些我用来尝试做出预测的样本图片。所有这些都显示为8级。
答案 0 :(得分:0)
我确实遇到了你所描述的问题。我也希望有人能分享他/她的想法。训练的测试图像是黑色底色,图形是白色。但是一旦我用我的数据预测模型,它总是显示错误的数字。我试图改变背景颜色,但它预测不对。训练后即使准确度也相当不错。图像是从网络资源中收集并由我自己绘制的,它清晰可读且符合条件。我在这里粘贴了我的预测代码:
import tensorflow as tf
import pandas as pd
from cnn_mnist import cnn_model_fn
import numpy as np
import cv2
# Predict the image
#mnist = tf.contrib.learn.datasets.load_dataset("mnist")
#train_labels = np.asarray(mnist.train.labels, dtype=np.int32)
#eval_data = mnist.test.images
model = cnn_model_fn
mnist_classifier = tf.estimator.Estimator(
model_fn=model, model_dir="/tmp/mnist_convnet_model")
im = cv2.imread("D:\\tf_exe_2\\img_9.jpg")
#im2 = np.reshape(im, [-1, 28, 28, 1])
pred_input_fn = tf.estimator.inputs.numpy_input_fn(
x={"x": np.float32(im)},
shuffle=False)
pred = list(mnist_classifier.predict(input_fn=pred_input_fn))
for el in pred:
print(el)
#print(next(pred))
结果是错误的。它应该在图像内容中为零。
==================== RESTART: D:\tf_exe_2\cnn_predict.py ====================
INFO:tensorflow:Using default config.
INFO:tensorflow:Using config: {'_keep_checkpoint_every_n_hours': 10000, '_service': None, '_task_id': 0, '_session_config': None, '_log_step_count_steps': 100, '_save_checkpoints_steps': None, '_save_summary_steps': 100, '_keep_checkpoint_max': 5, '_tf_random_seed': None, '_num_worker_replicas': 1, '_model_dir': '/tmp/mnist_convnet_model', '_num_ps_replicas': 0, '_task_type': 'worker', '_is_chief': True, '_save_checkpoints_secs': 600, '_cluster_spec': <tensorflow.python.training.server_lib.ClusterSpec object at 0x000001B2CD51E588>, '_master': ''}
INFO:tensorflow:Restoring parameters from /tmp/mnist_convnet_model\model.ckpt-21020
{'probabilities': array([0., 0., 1., 0., 0., 0., 0., 0., 0., 0.], dtype=float32), 'classes': 2}
{'probabilities': array([0., 0., 0., 0., 0., 0., 0., 0., 1., 0.], dtype=float32), 'classes': 8}
{'probabilities': array([0., 0., 0., 0., 0., 0., 0., 0., 1., 0.], dtype=float32), 'classes': 8}
>>>
所有培训步骤和配置都遵循Tensorflow教程。也许它会尝试继续修改训练参数,例如shuffle或其他配置?
这是cnn_mnist.py,与TensorFlow教程完全相同:
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
#imports
import numpy as np
import tensorflow as tf
tf.logging.set_verbosity(tf.logging.INFO)
def cnn_model_fn(features, labels, mode):
"""Model function for CNN"""
#Input Layer
input_layer = tf.reshape(features["x"], [-1,28,28,1])
#Convolutional Layer #1
conv1 = tf.layers.conv2d(
inputs = input_layer,
filters = 32,
kernel_size=[5,5],
padding = "same",
activation=tf.nn.relu)
#Pooling Layer #1
pool1 = tf.layers.max_pooling2d(inputs=conv1, pool_size=[2,2], strides=2)
#Convolutional Layer #2 and Pooling Layer #2
conv2 = tf.layers.conv2d(
inputs=pool1,
filters=64,
kernel_size=[5,5],
padding="same",
activation=tf.nn.relu)
pool2 = tf.layers.max_pooling2d(inputs=conv2, pool_size=[2,2], strides=2)
#Dense Layer
pool2_flat = tf.reshape(pool2, [-1,7*7*64])
dense = tf.layers.dense(inputs=pool2_flat, units=1024, activation=tf.nn.relu)
dropout = tf.layers.dropout(
inputs=dense, rate=0.4, training=mode == tf.estimator.ModeKeys.TRAIN)
#Logits Layer
logits = tf.layers.dense(inputs=dropout, units=10)
predictions = {
#Generate predictions (for PREDICT and EVAL mode)
"classes": tf.argmax(input=logits, axis=1),
#Add 'softmax_tensor' to the graph. It is used for PREDICT and by the
#'logging_hook'
"probabilities": tf.nn.softmax(logits, name="softmax_tensor")
}
if mode == tf.estimator.ModeKeys.PREDICT:
return tf.estimator.EstimatorSpec(mode=mode, predictions=predictions)
# Calculate Loss (for both TRAIN and EVAL modes
loss = tf.losses.sparse_softmax_cross_entropy(labels=labels, logits=logits)
# Configure the Training Op (for TRAIN mode)
if mode == tf.estimator.ModeKeys.TRAIN:
optimizer = tf.train.GradientDescentOptimizer(learning_rate=0.001)
train_op = optimizer.minimize(
loss=loss,
global_step=tf.train.get_global_step())
return tf.estimator.EstimatorSpec(mode=mode, loss=loss, train_op=train_op)
# Add evaluation metrics (for EVAL mode)
eval_metric_ops = {
"accuracy": tf.metrics.accuracy(
labels=labels, predictions=predictions["classes"])}
return tf.estimator.EstimatorSpec(
mode=mode, loss=loss,eval_metric_ops=eval_metric_ops)
def main(unused_argv):
#Load training and eval data
mnist = tf.contrib.learn.datasets.load_dataset("mnist")
train_data = mnist.train.images
train_labels = np.asarray(mnist.train.labels, dtype=np.int32)
eval_data = mnist.test.images
eval_labels = np.asarray(mnist.test.labels, dtype=np.int32)
#Create the Estimator
mnist_classifier = tf.estimator.Estimator(
model_fn=cnn_model_fn, model_dir="/tmp/mnist_convnet_model")
# Set up logging for predictions
tensor_to_log = {"probabilities": "softmax_tensor"}
logging_hook = tf.train.LoggingTensorHook(
tensors=tensor_to_log, every_n_iter=50)
# Train the model
train_input_fn = tf.estimator.inputs.numpy_input_fn(
x={"x": train_data},
y=train_labels,
batch_size=100,
num_epochs=None,
shuffle=True)
mnist_classifier.train(
input_fn=train_input_fn,
#original steps are 20000
steps=20000,
hooks=[logging_hook])
# Evaluate the model and print results
eval_input_fn = tf.estimator.inputs.numpy_input_fn(
x={"x": eval_data},
y=eval_labels,
num_epochs=1,
shuffle=False)
eval_results = mnist_classifier.evaluate(input_fn=eval_input_fn)
print(eval_results)
if __name__ == "__main__":
tf.app.run()