![](https://img-blog.csdnimg.cn/20201014180756925.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
Tensorflow
文章平均质量分 63
aprilvkuo
这个作者很懒,什么都没留下…
展开
-
What's the difference between tf.placeholder and tf.Variable
https://stackoverflow.com/questions/36693740/whats-the-difference-between-tf-placeholder-and-tf-variabledown vote The difference is that with tf.Variable you have to provide an initial value when you翻译 2017-11-30 01:02:00 · 184 阅读 · 0 评论 -
tensorflow 维度变化
https://www.tensorflow.org/api_guides/python/array_ops#Shapes_and_Shapingreshapelike numpysqueeze(removes dimensions of size 1 from the shape of a tensor)expand_dims(adds dimensions of size 1)原创 2017-11-30 09:20:50 · 1023 阅读 · 0 评论 -
TensorFlow 中生成Batch数据
TensorFlow 中生成Batch数据Reference: https://www.tensorflow.org/programmers_guide/datasets#simple_batching1. 利用sklearndef batch(self,STEPS=10000, BATCH=100): ss = ShuffleSplit(n_splits=STEPS, train_size原创 2017-12-01 14:53:19 · 4988 阅读 · 0 评论 -
Difference between Steps,Batch size,Epoch in Tensorflow
https://stackoverflow.com/questions/42816124/steps-vs-num-epochs-in-tensorflow-getting-started-tutorial Let’s start the opposite the order:1) Steps - number of times the training loop in your learning转载 2017-12-02 00:45:34 · 1893 阅读 · 0 评论 -
在TensorFlow使用RNN(读书笔记)
在TensorFlow使用RNNRNN 输入的是有多个时间点序列。tf.SequenceExample在tensorflow中有用来处理这样数据的数据交换格式( protocol buffer)。虽然也可以用python 或者 Numpy 的array,但是tf.SequenceExample有下面的优点。 优点: 简单,可以把数据分成多个TFRecord,每一个含有多个序列样例。而且可以支持Te原创 2017-11-24 00:16:42 · 970 阅读 · 0 评论