If I start with the wide data format, a 2d array of samples (rows) by sensor readings (columns), what is the right way to transform that to fit the requirements of this library?

Thank you very much, once again, for this great piece of software. Very much appreciated! I'm trying to use it with my data but unfortunately, I always get the following error if I attempt to fit my input with "parameters = fit(x_trainScaled)":

TypeError: No matching definition for argument type(s) array(float64, 2d, C), array(int32, 1d, C), array(int32, 1d, C), array(float32, 1d, C)

Here are some, probably, relevant characteristics of my input:

    print(x_trainScaled.shape)
    print(x_trainScaled.dtype)

returns:

(3000, 3000)
float64

// edit:

This is the whole traceback:

  File "minirocket\code\minirocket.py", line 130, in fit
    biases = _fit_biases(X, dilations, num_features_per_dilation, quantiles)
  File "\lib\site-packages\numba\dispatcher.py", line 500, in _explain_matching_error
    raise TypeError(msg)

Hello, I'm trying to figure out what minirocket expects as data on input. I keep on getting TypeError: No matching definition for argument type(s) pyobject, array(int32, 1d, C), array(int32, 1d, C), array(float32, 1d, C)

My data has following format:

timestamp,close
1619773130596,54559.47
1619773134938,54563.93
1619773139226,54554.23
1619773143564,54564.34

And I read it like this:

dataset = pd.read_csv(filename, usecols = [0, 1], header=0)
dataset = dataset.dropna()
dataset.columns = dataset.columns.to_series().apply(lambda x: x.strip())

hello, I watch the code about multivarible miniroket. I think the combine multi channels is not make sense for me. Conv(x) , x is channel 0 Conv(y), y is channel 1 when combine the channel, just become: Conv(x+y) why not, change the np.sum to np.prod. Conv(x*y)

hello, thanks for your excellent work. wmm, and I have a problem, I find the response in "starting with "wide" data", you say the data can be unlabeled, it depends on my task "(You don't need labels necessarily, depending on your task.)" and when I read your article or code readme, I notice that you mentioned the parameters in different data are same, right? (ok, I don't know if I understand right, and I can't find where is the latter information.) So my question is, could I apply your work on my unlabeled data? if it's true, how can I set the "Y_traing" in examples codes? thanks!

Thank you so much for making your work available! I have a quick question about the feature size. Looks like the minimum number of feature size is 84. Is there any harm in extracting 84 features and using only a subset them?

My setup is that I am using large dataset (10,000+) and I pass data as batches into model. I do not cache the data and run transform every time I pass data into model on every epoch. I run this same setup for both

minirocket.py with input shape (32768,99) and

minirocket_multivariate.py with input shape (32768,1,99) so the number of channel is 1.

I find that the minirocket_multivariate.py version runs significantly more slow on every transform() relative to minirocket.py.

Is there a potential bug in the code?

hi, for softmax.py, if the data is split into multiple chunks, then X_validation is only transformed for the first's chunk biases, as biases for different chunks are different, but the transform is only applied once.

if epoch == 0 and chunk_index == 0: # only run once <---

   parameters = fit(X_training, args["num_features"]) # returns: dilations, num_features_per_dilation, biases

   # transform validation data
   X_validation_transform = transform(X_validation, parameters)

would transforming the X_validation with each chunk's biases improve performance?

EDIT:

similarly for the latter part (where X_validation_transform is only normalised with mean and std values from the first chunk):

if epoch == 0 and chunk_index == 0:

                    # per-feature mean and standard deviation
                    f_mean = X_training_transform.mean(0)
                    f_std = X_training_transform.std(0) + 1e-8

                    # normalise validation features
                    X_validation_transform = (X_validation_transform - f_mean) / f_std
                    X_validation_transform = torch.FloatTensor(X_validation_transform)

when i use my data with minirocket in pycharm , had a problem with dataype, like: Traceback (most recent call last): File "E:/PycharmProjects/minirocket-main/code/traintest.py", line 46, in parameters = fit(X_training) File "E:\PycharmProjects\minirocket-main\code\minirocket.py", line 130, in fit biases = _fit_biases(X, dilations, num_features_per_dilation, quantiles) File "E:\ProgramData\Anaconda3\envs\deepl\lib\site-packages\numba\core\dispatcher.py", line 703, in _explain_matching_error raise TypeError(msg)TypeError: No matching definition for argument type(s) pyobject, array(int32, 1d, C), array(int32, 1d, C), array(float32, 1d, C) how can i work it

Dependencies

import numpy as np
from sklearn.linear_model import RidgeClassifier
from sklearn.pipeline import Pipeline
from sklearn.model_selection import GridSearchCV

from sktime.datasets import load_basic_motions
from sktime.transformations.panel.rocket import MiniRocketMultivariate`

Make train/test split and set up pipeline

X_train, y_train = load_basic_motions(split="train", return_X_y=True)

model = Pipeline([
    ('minirocket', MiniRocketMultivariate(random_state=42)), 
    ('ridge_clf', RidgeClassifier(random_state=42)),
])

Fit 1 model

model.fit(X_train, y_train) Works fine

Now do a gridsearch for alpha value

parameters = {
  'ridge_clf__alpha': [0.1, 1, 10],
}

model_cv = GridSearchCV(model, parameters)

model_cv.fit(X_train, y_train)

"RuntimeError: Cannot clone object MiniRocketMultivariate(random_state=42), as the constructor either does not set or modifies parameter random_state"

Hello,

The implementations for minirocket multivariate (both here and on sktime) mention that it is a naive extension of the univatiate version, but do not give any clearer explanation of what is actually happening under the hood. Looking directly at the source code for this version does not help that much either, as it is fairly hard to read.

Could you extend the documentation on the repository with a (coarse) description of how the algorithm was extended to handle multivariate data and/or add some comments to the source code in that regard?

Thanks!