get raw data numpy array

[PhD-GOAT (GOATLAB)]

• MNE version: 1.0.2
• operating system: Windows 10

import mne
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

raw = mne.io.read_raw_edf('C:/Users//SC4001E0-PSG.edf', preload=True)
print(raw)
print(raw.ch_names)

info = mne.create_info(ch_names=['EEG Fpz-Cz'],
                       ch_types=['eeg'],
                       sfreq=100)

raw1 = raw.pick_channels(['EEG Fpz-Cz'])
np_raw = mne.io.RawArray(raw1, info)
print(np_raw)

ValueError: All picks must be < n_channels (1), got 1

[richard (Richard Höchenberger)]

There is only one channel in the data, hence you cannot pick, as suggested by the error message.

[PhD-GOAT (GOATLAB)]

Right, but If erase this raw1 = raw.['EEG Fpz-Cz'], it says All picks must be < n_channels (7), got 7.

The raw data used consists of seven channels.

[richard (Richard Höchenberger)]

Please post the full code and complete error message (the complete traceback)

[PhD-GOAT (GOATLAB)]

import mne
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt

raw = mne.io.read_raw_edf('C:/Users//SC4001E0-PSG.edf', preload=True)
print(raw)
print(raw.ch_names)

info = mne.create_info(ch_names=['EEG Fpz-Cz'],
                       ch_types=['eeg'],
                       sfreq=100)

np_raw = mne.io.RawArray(raw, info)
print(np_raw)

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_11320/1799635321.py in <module>
     15                        sfreq=100)
     16 
---> 17 np_raw=mne.io.RawArray(raw, info)
     18 print(np_raw)

<decorator-gen-219> in __init__(self, data, info, first_samp, copy, verbose)

~\anaconda3\lib\site-packages\mne\io\array\array.py in __init__(self, data, info, first_samp, copy, verbose)
     55         _validate_type(info, 'info', 'info')
     56         _check_option('copy', copy, ('data', 'info', 'both', 'auto', None))
---> 57         dtype = np.complex128 if np.any(np.iscomplex(data)) else np.float64
     58         orig_data = data
     59         data = np.asanyarray(orig_data, dtype=dtype)

<__array_function__ internals> in iscomplex(*args, **kwargs)

~\anaconda3\lib\site-packages\numpy\lib\type_check.py in iscomplex(x)
    238 
    239     """
--> 240     ax = asanyarray(x)
    241     if issubclass(ax.dtype.type, _nx.complexfloating):
    242         return ax.imag != 0

~\anaconda3\lib\site-packages\numpy\core\_asarray.py in asanyarray(a, dtype, order, like)
    169         return _asanyarray_with_like(a, dtype=dtype, order=order, like=like)
    170 
--> 171     return array(a, dtype, copy=False, order=order, subok=True)
    172 
    173 

~\anaconda3\lib\site-packages\mne\io\base.py in __getitem__(self, item)
    807 
    808         """  # noqa: E501
--> 809         return self._getitem(item)
    810 
    811     def _getitem(self, item, return_times=True):

~\anaconda3\lib\site-packages\mne\io\base.py in _getitem(self, item, return_times)
    810 
    811     def _getitem(self, item, return_times=True):
--> 812         sel, start, stop = self._parse_get_set_params(item)
    813         if self.preload:
    814             data = self._data[sel, start:stop]

~\anaconda3\lib\site-packages\mne\io\base.py in _parse_get_set_params(self, item)
    742                                "and time)")
    743 
--> 744         sel = _picks_to_idx(self.info, item[0])
    745 
    746         if isinstance(item[1], slice):

~\anaconda3\lib\site-packages\mne\io\pick.py in _picks_to_idx(info, picks, none, exclude, allow_empty, with_ref_meg, return_kind)
   1067                          % (-n_chan, orig_picks))
   1068     if (picks >= n_chan).any():
-> 1069         raise ValueError('All picks must be < n_channels (%d), got %r'
   1070                          % (n_chan, orig_picks))
   1071     picks %= n_chan  # ensure positive

ValueError: All picks must be < n_channels (7), got 7

[richard (Richard Höchenberger)]

Does RawArray actually accept Raw? I thought it needs an array as input.

In any case, I don’t understand what you’re even trying to do here? Why do you load a raw file, just to take it apart and generate a RawArray manually from it? This doesn’t make any sense to me

[PhD-GOAT (GOATLAB)]

In order to use the function of yasa, any channel in edf must be converted to a numpy array.

[richard (Richard Höchenberger)]

Why don’t you just use raw.get_data() to get an array?

[PhD-GOAT (GOATLAB)]

Good. I asked because I couldn’t find raw.get_data with ctrl+F in api.

[drammock (Dan McCloy)]

rather than “using ctrl+f in the API” (not clear where exactly you were searching?) I suggest using the search box on our website. You’ll see that I was able to see several results when searching for raw.get_data; the second one is the most appropriate result but any of the first three would likely have told you what you needed to know: