Something wrong with the code when i use self-supervised mode

emadeldeen24 / TS-TCC

[IJCAI-21] "Time-Series Representation Learning via Temporal and Contextual Contrasting"

MIT License

339 stars 98 forks source link

Something wrong with the code when i use self-supervised mode #33

Closed kkizjw closed 2 months ago

kkizjw commented 2 months ago

when i use self-supervised mode, the code report "setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (5,) + inhomogeneous part." in the line of "warp = np.concatenate(np.random.permutation(splits)).ravel()" in "augmentation" .How should i change the code to run it successfully?

emadeldeen24 commented 2 months ago

Refer to this issue https://github.com/emadeldeen24/TS-TCC/issues/27