search

benkrasnow / LeadCrystal

Code and data from measurement of Lead (Pb) crystal leaching lead into beverages
14 stars 1 forks source link

Request - Raw Data #1

Open charliepuch opened 1 year ago

charliepuch commented 1 year ago

Hi Ben,

I recently watched your video on measuring lead concentration and found it inspiring as always. I really appreciated the detail you went into on the chemometric analysis you performed on the spectroscopic samples. I work/worked for a spectroscopy company for 10 years, and they have their own software package for viewing, processing, and building chemometric models using spectral data called NovaMath.

I wanted to take a shot at processing your data from your Lead experiment, so I downloaded the files form this repository and had a look at the files in the raw data folder. It looks like all the files in the raw data folder are multiple scans of the same sample, as they all look almost identical.

I was wondering if you were able to share the spectral data from the other samples here on Git Hub? I'd be happy to share the final model with you and the processing techniques I used.

If not, awesome video and experiment as always!

Charlie

benkrasnow commented 1 year ago

Charlie, it's true that I didn't save raw data for the whole experiment. The .txt files are just a few to show the structure of data from the spectrometer. The sample data, and calibration runs are in samplesession and calsession respectively. These contain raw data for the whole calibration run with named lead concentrations, and also sample raw data for the final acquisition (3 hour extraction after the 6day extraction and recovery period). You'll need octave or matlab to load the files, and will need to check out the .m code to see where everything is stored. I have to admit, I didn't initially intend to publish any of this, so it's pretty messy. Let me know if you have any questions.

charliepuch commented 1 year ago

Hi Ben,

Thanks so much for the reply. Ahh okay, I'll need to look into the sample and cal session files to see if they contain the raw spectral data I was interested in playing around with. If not, no worries at all.

I was curious to try different pre processing algorithms on the data such as derivatives or standard normal variate. These are common for normalizing AU data. Once processed I was then was curious about building a PLS model and see what kind of results it would achieve.

I really appreciate all the work you do on YouTube, your experiments keep getting more and more interesting.

Keep up the good work!

P.S. I might have some surplus optics gear I could send to you. I'll take a look in the storage room and if you'd be interested I'll send you a list of things I could send.

Charlie

On Sun, Jul 9, 2023, 1:29 PM benkrasnow @.***> wrote:

Charlie, it's true that I didn't save raw data for the whole experiment. The .txt files are just a few to show the structure of data from the spectrometer. The sample data, and calibration runs are in samplesession and calsession respectively. These contain raw data for the whole calibration run with named lead concentrations, and also sample raw data for the final acquisition (3 hour extraction after the 6day extraction and recovery period). You'll need octave or matlab to load the files, and will need to check out the .m code to see where everything is stored. I have to admit, I didn't initially intend to publish any of this, so it's pretty messy. Let me know if you have any questions.

— Reply to this email directly, view it on GitHub https://github.com/benkrasnow/LeadCrystal/issues/1#issuecomment-1627790601, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKLBOUBGC5WGOP7WKIQ7HLDXPL2CJANCNFSM6AAAAAA2DUC2S4 . You are receiving this because you authored the thread.Message ID: @.***>