koutoulakis / deep-learning-for-human-activity-recognition Goto Github PK

hey,

why did u go for ['S2-ADL3.dat', 'S2-ADL4.dat','S3-ADL3.dat', 'S3-ADL4.dat'] as testing set in the opportunity dataset?
why not replicate the challenge - which would be 'S2-ADL4','S2-ADL5','S3-ADL4','S3-ADL5'

besides: if I remember correctly, subject 4 data has induced noise.

Cheers

I got this error when i run the CNN file by gait dataset

segmenting signal...
Traceback (most recent call last):
File "C:\Users\ADMIN\gaitdap\ModelCreation\CNN\cnn1d.py", line 209, in
train_x, train_y = segment_dap(x_train,y_train,input_width)
File "C:\Users\ADMIN\gaitdap\ModelCreation\CNN\cnn1d.py", line 63, in segment_dap
if(int(len(x_train[start:end])) == window_size):
TypeError: slice indices must be integers or None or have an index method

please tell me how to solve it

Hello,
I believe you are appending the labels in the input data for the PAMAP2 dataset.
Lines 143 and 144 of the datareader.py file.
Cheers,
Enrique Marquez

I tried to run the CNN code but this error came
File "C:\Users\ADMIN\ana\envs\tensorflow\lib\site-packages\tensorflow\python\client\session.py", line 1104, in _run
% (np_val.shape, subfeed_t.name, str(subfeed_t.get_shape())))

ValueError: Cannot feed value of shape (64, 3) for Tensor 'Placeholder_4:0', which has shape '(?, 2)'