Hi James
Thank you for creating such a wonderful app. While I used the app, I indeed had some questions. I recorded the data in csv files and for each signal, I got 4 channels namely TP9, AF7 etc. I am trying to plot the data on Python using the average of these 4 channels. But what I got is not as smooth as the plot I got on the website. In addition, it's not scaled from 0-100 as well. Could you tell me what did you use for the plotting? Thank you in advance and have a good day
enquiry about the plot
Re: enquiry about the plot
You can just view the source code if you want. It's all in Javascript and I coded it to be readable. 
The absolutes values as read from the sensors are commonly in the {-1:+1} range. A lot of people found negative values confusing (assuming they had some significance when they do not), so they are converted (linear conversion) to the {0:100} range for display.
The Google Charts API has a "smoothing" option. You can toggle it on and off with the "Smooth Curve" checkbox.
Lastly to make the chart more accessible from lower powered devices, by default the data is grouped by minute, so you get a single data point on the chart with a per minute average value. This makes the chart load much faster. If you want to see all the data, you can check "All Data Points", which will group per second, and if you have recorded your CSV at 256Hz and really want to crash your computer with insane granularity, there's a hidden special chart here: https://mind-monitor.com/ChartMS.php , where the "All Data Points" button, actually gives you everything down to the millisecond.
The absolutes values as read from the sensors are commonly in the {-1:+1} range. A lot of people found negative values confusing (assuming they had some significance when they do not), so they are converted (linear conversion) to the {0:100} range for display.
The Google Charts API has a "smoothing" option. You can toggle it on and off with the "Smooth Curve" checkbox.
Lastly to make the chart more accessible from lower powered devices, by default the data is grouped by minute, so you get a single data point on the chart with a per minute average value. This makes the chart load much faster. If you want to see all the data, you can check "All Data Points", which will group per second, and if you have recorded your CSV at 256Hz and really want to crash your computer with insane granularity, there's a hidden special chart here: https://mind-monitor.com/ChartMS.php , where the "All Data Points" button, actually gives you everything down to the millisecond.
Re: enquiry about the plot
Hi James
Thank you for your answer, I have one more about concentration level detection. I really want to know how your algorithm determines if someone is focused or not. Could you please explain it in a more explicit way and hopefully I can get inspired.I have seen someone asked the similar question here, but I don't quite get it about tracking the gamma relative and "zero" at T+30 seconds, can you please explain it?
Thank you for your answer, I have one more about concentration level detection. I really want to know how your algorithm determines if someone is focused or not. Could you please explain it in a more explicit way and hopefully I can get inspired.I have seen someone asked the similar question here, but I don't quite get it about tracking the gamma relative and "zero" at T+30 seconds, can you please explain it?
Re: enquiry about the plot
It's a terrible algorithm. You don't want to copy it. It's just plotting Gamma relative with the zero point being whatever the value is after 30 seconds of having good sensor contact.