Problem with Augmentation

[sungsy]

I am currently trying to get the repository running. Soon as I try to run the self-supervised part, I run into the following error. I downloaded the datasets from the dataverse... I understand the error message and can follow it, but I am wondering why it does not work for me out of the box. Is there anything I am missing here?

> python main.py --training_mode self_supervised --selected_dataset HAR
=============================================
Dataset: HAR
Method:  TS-TCC
Mode:    self_supervised
=============================================
Traceback (most recent call last):
  File "/home/TS-TCC/main.py", line 85, in <module>
    train_dl, valid_dl, test_dl = data_generator(data_path, configs, training_mode)
  File "/home/TS-TCC/dataloader/dataloader.py", line 51, in data_generator
    train_dataset = Load_Dataset(train_dataset, configs, training_mode)
  File "/home/TS-TCC/dataloader/dataloader.py", line 33, in __init__
    self.aug1, self.aug2 = DataTransform(self.x_data, config)
  File "/home/TS-TCC/dataloader/augmentations.py", line 8, in DataTransform
    strong_aug = jitter(permutation(sample, max_segments=config.augmentation.max_seg), config.augmentation.jitter_ratio)
  File "/home/TS-TCC/dataloader/augmentations.py", line 42, in permutation
    warp = np.concatenate(np.random.permutation(splits)).ravel()
  File "numpy/random/mtrand.pyx", line 4720, in numpy.random.mtrand.RandomState.permutation
ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (3,) + inhomogeneous part.

[emadeldeen24]

It looks like an issue with the data dimensionality or format. What is the source of the data?

[sungsy]

I downloaded the HAR-dataset from here and placed it inside data/HAR/(test.pt, train.pt, val.pt). When I run python main.py --experiment_description exp1 --run_description run_1 --seed 123 --training_mode self_supervised --selected_dataset HAR I get the error above.

The training data has after extracting from dictionary a shape of torch.Size([5881, 9, 128]), torch.Size([5881]) for X and y.

[emadeldeen24]

I'm not sure, but I guess it could be due to different Numpy versions. Unfortunately, I haven't recorded which version I used. However, you can try downgrading and see if the problem persists.

[sungsy]

Issue could be solved by downgrading to numpy 1.21.0. Not sure if newer version could also work.
Following warning shown now (also for numpy 1.20.3):

/home/TS-TCC/dataloader/augmentations.py:42: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray.
  warp = np.concatenate(np.random.permutation(splits)).ravel()
<__array_function__ internals>:5: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray.
Data loaded ...
Training started ....
/home/TS-TCC/models/TC.py:52: UserWarning: Implicit dimension choice for log_softmax has been deprecated. Change the call to include dim=X as an argument.
  nce += torch.sum(torch.diag(self.lsoftmax(total)))`