LSTM-Human-Activity-Recognition-master

代码说明：

说明：  与经典的方法相比，使用具有长时间记忆细胞的递归神经网络(RNN)不需要或几乎不需要特征工程。数据可以直接输入到神经网络中，神经网络就像一个黑匣子，可以正确地对问题进行建模。其他研究在活动识别数据集上可以使用大量的特征工程，这是一种与经典数据科学技术相结合的信号处理方法。这里的方法在数据预处理的数量方面非常简单(Compared with the classical methods, the recursive neural network (RNN) with long-term memory cells does not need or almost need feature engineering. Data can be directly input into the neural network, which acts as a black box and can correctly model the problem. Other research can use a lot of Feature Engineering on activity recognition data sets, which is a signal processing method combined with classical data science and technology. The method here is very simple in terms of the number of data preprocessing)

文件列表：

LSTM-Human-Activity-Recognition-master, 0 , 2019-04-16
LSTM-Human-Activity-Recognition-master\.gitignore, 24 , 2019-04-16
LSTM-Human-Activity-Recognition-master\LICENSE, 1086 , 2019-04-16
LSTM-Human-Activity-Recognition-master\LSTM.ipynb, 213291 , 2019-04-16
LSTM-Human-Activity-Recognition-master\LSTM_files, 0 , 2019-04-16
LSTM-Human-Activity-Recognition-master\LSTM_files\LSTM_16_0.png, 77480 , 2019-04-16
LSTM-Human-Activity-Recognition-master\LSTM_files\LSTM_18_1.png, 43286 , 2019-04-16
LSTM-Human-Activity-Recognition-master\README.md, 30154 , 2019-04-16
LSTM-Human-Activity-Recognition-master\data, 0 , 2019-04-16
LSTM-Human-Activity-Recognition-master\data\.gitignore, 33 , 2019-04-16
LSTM-Human-Activity-Recognition-master\data\download_dataset.py, 914 , 2019-04-16
LSTM-Human-Activity-Recognition-master\data\source.txt, 2068 , 2019-04-16