What input does Online feature MiniRocket expect?

[jerronl]

Because I can't run the functin get_UCR_data issue#818 I can't figuure out what input expected by MiniRocket . I tried following code and it failed

print(X.shape,y.shape)

tfms = [None, TSClassification()]
batch_tfms = TSStandardize(by_sample=True)
dls = get_ts_dls(X, y, splits=splits, tfms=tfms, batch_tfms=batch_tfms)
from tsai.models.MINIROCKET_Pytorch import *
from tsai.models.utils import *
model = build_ts_model(MiniRocket, dls=dls)
print(model)
learn = Learner(dls, model, metrics=accuracy, cbs=ShowGraph())
learn.lr_find()

The error is after the loss_func call with blank result of irs and losses:

(8113, 21, 6) (8113,)
MiniRocket(
  (backbone): MiniRocketFeatures()
  (head): Sequential(
    (0): Flatten(start_dim=1, end_dim=-1)
    (1): BatchNorm1d(9996, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (2): Linear(in_features=9996, out_features=22, bias=True)
  )
)

---------------------------------------------------------------------------

IndexError                                Traceback (most recent call last)

[<ipython-input-15-f4f134341166>](https://localhost:8080/#) in <cell line: 6>()
      4 print(model)
      5 learn = Learner(dls, model, metrics=accuracy, cbs=ShowGraph())
----> 6 learn.lr_find()

1 frames

[/usr/local/lib/python3.10/dist-packages/fastai/callback/schedule.py](https://localhost:8080/#) in lr_find(self, start_lr, end_lr, num_it, stop_div, show_plot, suggest_funcs)
    302         for func in tuplify(suggest_funcs):
    303             nms.append(func.__name__ if not isinstance(func, partial) else func.func.__name__) # deal with partials
--> 304             _suggestions.append(func(lrs, losses, num_it))
    305 
    306         SuggestedLRs = collections.namedtuple('SuggestedLRs', nms)

[/usr/local/lib/python3.10/dist-packages/fastai/callback/schedule.py](https://localhost:8080/#) in valley(lrs, losses, num_it)
    229     idx = max_start + int(sections) + int(sections/2)
    230 
--> 231     return float(lrs[idx]), (float(lrs[idx]), losses[idx])
    232 
    233 # %% ../../nbs/14_callback.schedule.ipynb 81

IndexError: index 0 is out of bounds for dimension 0 with size 0

can you tell me what inputs are expected?

[oguiza]

Hi @jerronl,
Please try it again. The downloading data issue is fixed now in the github version. I'll soon create a new pypi / conda version.