search

hfawaz / aaltd18

Data augmentation using synthetic data for time series classification with deep residual networks
GNU General Public License v3.0
183 stars 42 forks source link

ValueError: Buffer has wrong number of dimensions (expected 2, got 1) #11

Open sadransh opened 3 years ago

sadransh commented 3 years ago

Thanks for sharing this work.

Is there any multivariate support? when Try to augment my data ( w/ shape (samples, length, n_features) ) I see the library is passing (samples, n_features) to dist_fun which results in this error 

aaltd18/dba.py in calculate_dist_matrix(tseries, dist_fun, dist_fun_params)
     10         for j in range(i+1,N):
     11             y = tseries[j]
---> 12             dist = dist_fun(x,y,**dist_fun_params)[0]
     13             # because dtw returns the sqrt
     14             dist = dist*dist
dtw.pyx in dtw.dynamic_time_warping()
ValueError: Buffer has wrong number of dimensions (expected 2, got 1)

what is the correct input type? (x and y both having shape (1,)