FFT is a methods for computing the DFT efficiently. In view of the importance of the DFT in various digital signal processing applications, such as linear filtering(线性滤波), correlation analysis（相关分析）, and spectrum analysis（谱分析）, its efficient computation is a topic that has received considerable attention by many mathematicians, engineers, and applied scientists.
Quantization error is the difference between the analog signal and the closest available digital value at each sampling instant from A/D converter. Quantization error also introduces noise,to the sample signal.
Here is a function commonly used: tf.nn.sparse_softmax_cross_entropy_with_logits(labels,logits) The labels is with type tf.int32 or tf.int64 and it’s rank is one less than logits. e.g: labels is in the shape of [d1,d2,…,dr], then the shape of logits should be [d1,d2,…,dr,n_class]. And the function return the tensor with the same shape of labels
My web
How to use Markdown
Markdown guidance and reference