glob – Filename pattern matching
tf.gfile.Glob
tf.gfile.Glob(filename)
Returns a list of files that match the given pattern(s).
Args:
- filename: string or iterable of strings. The glob pattern(s).
Returns:
- A list of strings containing filenames that match the given pattern(s).
Reference:
Example Data
The examples below assume the following test files are present in the current working directory:
1 | $ python glob_maketestdata.py |
Note Use glob_maketestdata.py in the sample code to create these files if you want to run the examples.
Wildcards
An asterisk (*) matches zero or more characters in a segment of a name. For example, dir/*.
1 | import glob |
The pattern matches every pathname (file or directory) in the directory dir, without recursing further into subdirectories.
1 | $ python glob_asterisk.py |
To list files in a subdirectory, you must include the subdirectory in the pattern:
1 | import glob |
The first case above lists the subdirectory name explicitly, while the second case depends on a wildcard to find the directory.
1 | $ python glob_subdir.py |
The results, in this case, are the same. If there was another subdirectory, the wildcard would match both subdirectories and include the filenames from both.
Single Character Wildcard
The other wildcard character supported is the question mark (?). It matches any single character in that position in the name. For example,
1 | import glob |
Matches all of the filenames which begin with “file”, have one more character of any type, then end with ”.txt”.
1 | $ python glob_question.py |
Reference
Tensorflow slim (TF-Slim)
Six 库
six是一个专门用来兼容 Python 2 和 Python 3 的库，用来使得代码同时在 Python 2 和 Python 3 上兼容。只需要在代码前加一句就可以了。
1 | import six |
tf.app.flags.FLAGS
tf.app.flags.FLAGS 可用来传递参数，
1 | import tensorflow as tf |
或者
1 | import tensorflow as tf |
tf.cast(x, dtype, name=None)
此函数是类型转换函数
参数：
- x：输入
- dtype：转换目标类型
- name：名称
- 返回：Tensor
例：
1 | # tensor `a` is [1.8, 2.2], dtype=tf.float |
tf.squeeze(
input,
axis=None,
name=None,
squeeze_dims=None
)1
2
3
4
去掉维数为1的维度。
例：
‘t’ is a tensor of shape [1, 2, 1, 3, 1, 1]
tf.shape(tf.squeeze(t)) # [2, 3]1
2
也可以指定去掉哪个维度：
‘t’ is a tensor of shape [1, 2, 1, 3, 1, 1]
tf.shape(tf.squeeze(t, [2, 4])) # [1, 2, 3, 1]1
2
3
4
5
6
7
8
9
10
11
12
## tf.where(a,b,c)
tf.where(a,b,c)函数：
功能：
当a输出结果为true时，tf.where(a,b,c)函数会选择b值输出。
当a输出结果为false时，tf.where(a,b,c)函数会选择c值输出。
例子：
import tensorflow as tf
v1=tf.constant([1.0,2.0,3.0,4.0])
v2=tf.constant([4.0,3.0,2.0,1.0])
with tf.Session() as sess:
init=tf.global_variables_initializer()
sess.run(init)
print(sess.run(tf.greater(v1,v2)))
print(sess.run(tf.where(tf.greater(v1,v2),v1,v2)))1
2
结果：
[False False True True]
[ 4. 3. 3. 4.]1
2
3
4
5
6
**转载自**
[https://blog.csdn.net/william_hehe/article/details/78636624](https://blog.csdn.net/william_hehe/article/details/78636624)
## tf.gather
gather(
params,
indices,
validate_indices=None,
name=None
)1
2
3
4
5
6
7
8
9
10
11
12
13
14
tf.gather(等待被取元素的张量，索引)
tf.gather根据索引，从输入张量中依次取元素，构成一个新的张量。索引的维度可以小于张量的维度。这时，取张量元素时，会把相应的低维当作一个整体取出来。
例如
假设输入张量 [[1,2,3],[4,5,6],[7,8,9]] 是个二维的。如果只给一个一维索引0. 它就把 [1,2,3] 整体取出。如果给两个一维索引，0和1，它就形成 [[1,2,3],[4,5,6]]
**转载自**
[https://zhuanlan.zhihu.com/p/34578921](https://zhuanlan.zhihu.com/p/34578921)
## tf.zeros_like
zeros_like(
tensor,
dtype=None,
name=None,
optimize=True
)1
2
3
4
Creates a tensor with all elements set to zero.
For example:
tensor = tf.constant([[1, 2, 3], [4, 5, 6]])
tf.zeros_like(tensor) # [[0, 0, 0], [0, 0, 0]]1
2
3
4
5
6
7
## tf.reshape
tf.reshape(tensor,shape, name=None)
函数的作用是将tensor变换为参数shape的形式。
其中shape为一个列表形式，特殊的一点是列表中可以存在-1。-1代表的含义是不用我们自己指定这一维的大小，函数会自动计算，但列表中只能存在一个-1。（当然如果存在多个-1，就是一个存在多解的方程了）
x = [[1 2 3]
[5 6 7]]
x_f = tf.reshape(x,(1,-1)) # x_f = [[1 2 3 5 6 7]]1
2
## tf.tile
tf.tile(
input,
multiples,
name=None
)1
2
3
4
This operation creates a new tensor by replicating input multiples times. The output tensor's i'th dimension has `input.dims(i) * multiples[i]` elements, and the values of input are replicated `multiples[i]` times along the 'i'th dimension. For example, tiling `[a b c d]` by `[2]` produces `[a b c d a b c d]`.
## tf.Session
import tensorflow as tf
Build a graph.
a = tf.constant(5.0)
b = tf.constant(6.0)
c = a * b
Launch the graph in a session.
sess = tf.Session()
Evaluate the tensor c.
print(sess.run(c))
import tensorflow as tf
a = tf.constant(5.0)
b = tf.constant(6.0)
c = a * b
with tf.Session():
We can also use ‘c.eval()’ here.
print(c.eval())1
2
3
4
5
## `tf.clip_by_value`
`tf.clip_by_value(A, min, max)`：输入一个张量A，把A中的每一个元素的值都压缩在min和max之间。小于min的让它等于min，大于max的元素的值等于max。
例如：
[python] view plain copy
import tensorflow as tf;
import numpy as np;
A = np.array([[1,1,2,4], [3,4,8,5]])
with tf.Session() as sess:
print sess.run(tf.clip_by_value(A, 2, 5))1
2
输出：
[[2 2 2 4]
[3 4 5 5]]`