# read\_dig\_dat() does not work **URL:** \ **Category:** Support & Discussions\ **Tags:** preprocessing, eeg, montage\ **Created:** [July 31, 2021, 2:33pm UTC](https://mne.discourse.group/t/read-dig-dat-does-not-work/3477 "2021-07-31T14:33:06Z")\ **Posts on this page:** 9\ **Page:** 1
**Author:** ![leosunpsy](https://avatars.discourse-cdn.com/v4/letter/l/5f8ce5/32.png) [@leosunpsy](https://mne.discourse.group/u/leosunpsy)\ **Post date:** [July 31, 2021, 2:33pm UTC](https://mne.discourse.group/t/read-dig-dat-does-not-work/3477/1 "2021-07-31T14:33:06Z")
- MNE-Python version: 0.23.0 - operating system: Windows 10 `montage = mne.channels.read_dig_dat(os.path.join(dataDir,'SynAmps2_Quik_Cap64.DAT'))` When i used the read\_dig\_dat() to read the SynAmps2\_Quik\_Cap64.DAT, this function did not work. I looked over the source code and found that the code handles five columns of data, but Compumedics only provides four columns for Cap64. Could anyone give me some advice on how to solve this problem? ---
**Author:** ![richard](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/richard/32/15_2.png) [@richard](https://mne.discourse.group/u/richard)\ **Post date:** [July 31, 2021, 3:02pm UTC](https://mne.discourse.group/t/read-dig-dat-does-not-work/3477/2 "2021-07-31T15:02:34Z")
Hello @leosunpsy and welcome to the forum! Could you share the problematic file so the developers can have a look? Thanks! ---
**Author:** ![leosunpsy](https://avatars.discourse-cdn.com/v4/letter/l/5f8ce5/32.png) [@leosunpsy](https://mne.discourse.group/u/leosunpsy)\ **Post date:** [July 31, 2021, 3:42pm UTC](https://mne.discourse.group/t/read-dig-dat-does-not-work/3477/3 "2021-07-31T15:42:21Z")
Hi, @richard. The SynAmps2\_Quik\_Cap64.DAT file can be download at [here](https://compumedicsneuroscan.com/scan-acquire-configuration-files/). I do not know how to directly upload files in the forum, please forgive me! Thanks for your reply. ---
**Author:** ![richard](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/richard/32/15_2.png) [@richard](https://mne.discourse.group/u/richard)\ **Post date:** [July 31, 2021, 4:15pm UTC](https://mne.discourse.group/t/read-dig-dat-does-not-work/3477/4 "2021-07-31T16:15:31Z")
Thank you! The format of this file is currently not supported by MNE-Python. It would be possible (and not too difficult) to add support for it, but somebody would have to do it. I’d be willing to give it a shot personally, but I’ll be busy for the next couple of days. If you can wait that long, I’d be glad to help. In the meantime, perhaps you could simply use our built-in 10-20 standard montage? It won’t give you the **precise** locations, but it might be good enough to move on with your sensor-level analysis? To do that, you can call ```python raw.set_montage(montage='standard_1020') ``` Best wishes, Richard ---
**Author:** ![leosunpsy](https://avatars.discourse-cdn.com/v4/letter/l/5f8ce5/32.png) [@leosunpsy](https://mne.discourse.group/u/leosunpsy)\ **Post date:** [August 1, 2021, 1:37am UTC](https://mne.discourse.group/t/read-dig-dat-does-not-work/3477/5 "2021-08-01T01:37:32Z")
@richard OK Thanks for your advice. ---
**Author:** ![leosunpsy](https://avatars.discourse-cdn.com/v4/letter/l/5f8ce5/32.png) [@leosunpsy](https://mne.discourse.group/u/leosunpsy)\ **Post date:** [August 1, 2021, 2:49am UTC](https://mne.discourse.group/t/read-dig-dat-does-not-work/3477/6 "2021-08-01T02:49:12Z")
Hi @richard, I have solved the problem. The modified code of the read\_dig\_dat function is as follows: ```python def read_dig_dat(fname): r"""Read electrode positions from a ``*.dat`` file. .. Warning:: This function was implemented based on ``*.dat`` files available from `Compumedics `__ and might not work as expected with novel files. If it does not read your files correctly please contact the mne-python developers. Parameters ---------- fname : path-like File from which to read electrode locations. Returns ------- montage : DigMontage The montage. See Also -------- read_dig_captrak read_dig_dat read_dig_egi read_dig_fif read_dig_hpts read_dig_polhemus_isotrak make_dig_montage Notes ----- ``*.dat`` files are plain text files and can be inspected and amended with a plain text editor. """ from ._standard_montage_utils import _check_dupes_odict fname = _check_fname(fname, overwrite='read', must_exist=True) with open(fname, 'r') as fid: lines = fid.readlines() ch_names, poss = list(), list() nasion = lpa = rpa = None for i, line in enumerate(lines): items = line.split() if not items: continue elif not (len(items) == 4 or len(items) == 5): raise ValueError( "Error reading %s, line %s has unexpected number of entries:\n" "%s" % (fname, i, line.rstrip())) if len(items) == 5: num = items[1] if num == '67': continue # centroid pos = np.array([float(item) for item in items[2:]]) if num == '78': nasion = pos elif num == '76': lpa = pos elif num == '82': rpa = pos else: ch_names.append(items[0]) poss.append(pos) elif len(items) == 4: label = items[0] if label == 'Centroid': continue # centroid pos = np.array([float(item) for item in items[1:]]) if label == 'Nasion': nasion = pos elif label == 'Left': lpa = pos elif label == 'Right': rpa = pos else: ch_names.append(items[0]) poss.append(pos) electrodes = _check_dupes_odict(ch_names, poss) return make_dig_montage(electrodes, nasion, lpa, rpa) ``` ---
**Author:** ![richard](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/richard/32/15_2.png) [@richard](https://mne.discourse.group/u/richard)\ **Post date:** [August 1, 2021, 6:28am UTC](https://mne.discourse.group/t/read-dig-dat-does-not-work/3477/7 "2021-08-01T06:28:56Z")
Nice! Would you like to try and make a pull request so your improvement can be officially included in the next release of MNE-Python? ---
**Author:** ![leosunpsy](https://avatars.discourse-cdn.com/v4/letter/l/5f8ce5/32.png) [@leosunpsy](https://mne.discourse.group/u/leosunpsy)\ **Post date:** [August 1, 2021, 7:16am UTC](https://mne.discourse.group/t/read-dig-dat-does-not-work/3477/8 "2021-08-01T07:16:34Z")
@richard I have made a pull request. ✌ ---
**Author:** ![richard](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/richard/32/15_2.png) [@richard](https://mne.discourse.group/u/richard)\ **Post date:** [August 1, 2021, 8:43am UTC](https://mne.discourse.group/t/read-dig-dat-does-not-work/3477/9 "2021-08-01T08:43:48Z")
Thank you! We will review the PR soon. It may take a few small adjustments until it can be merged. For reference, the PR can be found here: > > > Fixes #9631 > The mne.channels.read\_dig\_dat() function does not support the dat …file of neuroscan 64 channels. I have solved the problem. The modified code of the read\_dig\_dat function is as follows: > \`\`\` > def read\_dig\_dat(fname): > r"""Read electrode positions from a \`\`\*.dat\`\` file. > > .. Warning:: > This function was implemented based on \`\`\*.dat\`\` files available from > \`Compumedics \ configuration-files/\>\`\_\_ and might not work as expected with novel > files. If it does not read your files correctly please contact the > mne-python developers. > > Parameters > ---------- > fname : path-like > File from which to read electrode locations. > > Returns > ------- > montage : DigMontage > The montage. > > See Also > -------- > read\_dig\_captrak > read\_dig\_dat > read\_dig\_egi > read\_dig\_fif > read\_dig\_hpts > read\_dig\_polhemus\_isotrak > make\_dig\_montage > > Notes > ----- > \`\`\*.dat\`\` files are plain text files and can be inspected and amended with > a plain text editor. > """ > from .\_standard\_montage\_utils import \_check\_dupes\_odict > fname = \_check\_fname(fname, overwrite='read', must\_exist=True) > > with open(fname, 'r') as fid: > lines = fid.readlines() > > ch\_names, poss = list(), list() > nasion = lpa = rpa = None > for i, line in enumerate(lines): > items = line.split() > if not items: > continue > elif not (len(items) == 4 or len(items) == 5): > raise ValueError( > "Error reading %s, line %s has unexpected number of entries:\\n" > "%s" % (fname, i, line.rstrip())) > > if len(items) == 5: > num = items\[1\] > if num == '67': > continue # centroid > pos = np.array(\[float(item) for item in items\[2:\]\]) > if num == '78': > nasion = pos > elif num == '76': > lpa = pos > elif num == '82': > rpa = pos > else: > ch\_names.append(items\[0\]) > poss.append(pos) > elif len(items) == 4: > label = items\[0\] > if label == 'Centroid': > continue # centroid > pos = np.array(\[float(item) for item in items\[1:\]\]) > if label == 'Nasion': > nasion = pos > elif label == 'Left': > lpa = pos > elif label == 'Right': > rpa = pos > else: > ch\_names.append(items\[0\]) > poss.append(pos) > > electrodes = \_check\_dupes\_odict(ch\_names, poss) > return make\_dig\_montage(electrodes, nasion, lpa, rpa)