Research Article
Air-CSL: Chinese Sign Language Recognition Based on the Commercial WiFi Devices
Algorithm 1
Overall framework of Air-CSL.
Input: | Output: The loss value | Initialize the model parameters, the hidden layer dimension , the training rounds , the number , training loss value | 1: | 2: for to do | 3: use to update the reset door , the update door , and the hidden state | 4: forget unimportant information, update | 5: End for | 6: pass the data to the tensor | 7: for to ; | 8: if the current the predicted loss | 9: save model parameters | 10: and | 11: else if | 12: output the training rounds and | 13: End if | 14: End for | 15: Return |
|