MiniRocketPlus ValueError: Expected input batch_size (256) to match target batch_size (512)

[Zwayeh]

I am having an issue getting MiniRocketPlus to work with multivariant data.

#Create dataset
X, y = df2xy(PDCTAC1, target_col='target', feat_col='feature', sample_col='sample')
#Check size
test_eq(X.shape, (600, 2, 6541))
test_eq(y.shape, (600, 2))
#Generate train/valid splits
splits = get_splits(y, valid_size=.2, stratify=True, random_state=23, shuffle=True)

#Create dataloaders and attempt to train MiniRocketPlus
tfms = [None, TSClassification()]
batch_tfms = TSStandardize()
dls = get_ts_dls(X, y, splits=splits, tfms=tfms, batch_tfms=batch_tfms, bs=256)
learn = ts_learner(dls, MiniRocketPlus, kernel_size=7, metrics=accuracy)
learn.fit_one_cycle(1, 1e-2)

When y has more than one dimension, I receive an error: _ValueError: Expected input batch_size (256) to match target batchsize (512).

I can see the values from the error are created by the 'bs' variable in my dls dataloader, and this value multiplied by the number of features in my dataset.

[oguiza]

Please, do not post issues that have been previously discussed by email.