Dr. jonny phd on Nostr: nprofile1q…hcvwk Well you already put your interpretation on in it in choosing how ...
nprofile1qy2hwumn8ghj7un9d3shjtnddaehgu3wwp6kyqpq9d9p04u4xfysdy92fycw947jrca3xve2gnsauysshzewxvmz8dmsxhcvwk (nprofile…cvwk)
Well you already put your interpretation on in it in choosing how to plot it, so there isn't really a way to get away from that. Think about the way you read plots normally, unless you were trying to do some forensic analysis of the data you wouldn't be trying to recover the literal values, but would be looking for trends, tendencies, key points like where a curve peaks or changes slope. You tell the meaning of the data in the main text, and then the shape of the data gives some qualitative description of how the data feels - is there a gigantic difference between the training conditions or just a marginal one, you talk about performance at different sequence lengths, what are those lengths?
Ofc I am a sighted person and dont know what is best to do, so I'm just saying what I would do, idk if it's the best thing
Well you already put your interpretation on in it in choosing how to plot it, so there isn't really a way to get away from that. Think about the way you read plots normally, unless you were trying to do some forensic analysis of the data you wouldn't be trying to recover the literal values, but would be looking for trends, tendencies, key points like where a curve peaks or changes slope. You tell the meaning of the data in the main text, and then the shape of the data gives some qualitative description of how the data feels - is there a gigantic difference between the training conditions or just a marginal one, you talk about performance at different sequence lengths, what are those lengths?
Ofc I am a sighted person and dont know what is best to do, so I'm just saying what I would do, idk if it's the best thing