How can I solve EEG preprocessing problem about ICA?

The following error occurred when I run ICA. How can I solve it?

Beginning ICA training ... first training step may be slow ...

Lowering learning rate to 0.0009 and starting again.

Lowering learning rate to 0.00081 and starting again.

Lowering learning rate to 0.000729 and starting again.

step 1 - lrate 0.000729, wchange 154.68358687, angledelta 0.0 deg

step 2 - lrate 0.000729, wchange 51.37931089, angledelta 0.0 deg

Lowering learning rate to 0.0006561 and starting again.

Lowering learning rate to 0.00059049 and starting again.

step 1 - lrate 0.000590, wchange 150.12111179, angledelta 0.0 deg

Lowering learning rate to 0.000531441 and starting again.

step 1 - lrate 0.000531, wchange 143.41536306, angledelta 0.0 deg

Lowering learning rate to 0.000478297 and starting again.

step 1 - lrate 0.000478, wchange 132.83888872, angledelta 0.0 deg

step 2 - lrate 0.000478, wchange 33.55226181, angledelta 0.0 deg

Lowering learning rate to 0.000430467 and starting again.

step 1 - lrate 0.000430, wchange 120.15341119, angledelta 0.0 deg

step 2 - lrate 0.000430, wchange 34.18294986, angledelta 0.0 deg

step 3 - lrate 0.000430, wchange 32.45455678, angledelta 88.7 deg

Lowering learning rate to 0.000379672 and starting again.

step 1 - lrate 0.000380, wchange 107.10216681, angledelta 0.0 deg

step 2 - lrate 0.000380, wchange 32.01741739, angledelta 0.0 deg

step 3 - lrate 0.000380, wchange 33.11164277, angledelta 91.3 deg

step 4 - lrate 0.000372, wchange 31.51202714, angledelta 116.4 deg

step 5 - lrate 0.000365, wchange 33.58354390, angledelta 113.6 deg

step 6 - lrate 0.000357, wchange 32.87646615, angledelta 113.8 deg

Lowering learning rate to 0.000315178 and starting again.

step 1 - lrate 0.000315, wchange 84.16497884, angledelta 0.0 deg

step 2 - lrate 0.000315, wchange 28.46690639, angledelta 0.0 deg

Lowering learning rate to 0.00028366 and starting again.

step 1 - lrate 0.000284, wchange 80.39161447, angledelta 0.0 deg

step 2 - lrate 0.000284, wchange 28.03043140, angledelta 0.0 deg

In addition, there are my steps before ICA.

1. Downsample to 200Hz;
2. Filter: 1-45Hz;
3. Remove bad channels and artifacts using clean_rawdata();
4. Interpolate all the removed channels;
5. Re-reference the data to average

I am not sure whether the preprocessing is make sense? And how can I solve the ICA problem?

My computer version is Windows 10. Matlab version is R2020a. EEGLAB version is 2020.0.

Thank you very much.

[ke]

你好 你的问题解决了吗 我也遇到了相同的问题