Developer Terms and Conditions › The Development › Molecular Sensing Models › Classification model not working despite high correlation › Reply To: Classification model not working despite high correlation
Thanks Ayelet, Below are answers to your questions.
1. Have you scanned new samples in different temperatures? (specifically different temperatures than measured when collecting the database). I SCANNED THE COLLECTION SAMPLES AND TEST SAMPLES UNDER THE SAME CONDITION. THE TEMPERATURE MIGHT BE SLIGHTLY DIFFERENT AS I DID NOT TRY TO CONTROL THE TEMPERATURE EXACTLY.
2. Have you used actual different samples or scanned multiple scans for identical samples and referred them as new samples? I USED THE SAME PHYSICAL SAMPLE FOR 3 DATA COLLECTION SAMPLES (THE PHYSICAL SAMPLE IS JUST POSITIONED DIFFERENTLY IN EACH DATA COLLECTION SAMPLE). SO IN MY DATA COLLECTION, I HAVE 48 SAMPLES. THAT’S ACTUALLY FROM 16 DIFFERENT PHYSICAL SAMPLES.
3. Did you use the solid sample holder? I USED THE SOLID SAMPLE HOLDER. I’M ATTACHING PICTURES. I FIRST CRUSHED THE RICE SAMPLES INTO POWDERS AND FILLED THE SOLID SAMPLE HOLDER WITH 4 GRAMS OF POWDER.