Searching from 2022-11-20 00:00:00 to 2022-11-20 23:59:59.999999.
Query completed in 0.39 seconds
[2022-11-20 01:48:18] --> lukedashjr has joined the channel
[2022-11-20 01:48:55] <-- luke-jr has quit (Ping timeout: 268 seconds)
[2022-11-20 01:51:56] -!- lukedashjr changed nick to luke-jr
[2022-11-20 01:52:02] <-- Guest1818 has quit (Ping timeout: 480 seconds)
[2022-11-20 02:15:15] --> Luke-Jr has joined the channel
[2022-11-20 23:38:26] <Lucifer_arma> When analyzing accuracy over a large dataset, can you average the accuracies together?
[2022-11-20 23:38:59] <Lucifer_arma> The specific case here is that I've been training a bunch of different AIs to analyze stock price history, and I generate an overall accuracy of predictions for each individual stock
[2022-11-20 23:39:25] <Lucifer_arma> I want to gauge the accuracy of the bot over many stocks, so I want to average the accuracy of each prediction step over all of the stocks the bot studied
[2022-11-20 23:40:39] <Lucifer_arma> Each stock history dataset is different, because the stock itself could come from different industries, has different lengths of data (some companies haven't been publicly traded for long and may only have a few months history, while others have been trading for 20+ years and the dataset only goes back a maximum of 20 years)
[2022-11-20 23:41:19] <Lucifer_arma> Arguably, that makes each dataset unique, and so trying to combine them is wrong, but I'm trying to measure the general purpose accuracy of the bot in predicting any random stock I give it
[2022-11-20 23:42:07] <Lucifer_arma> For reference, I'm trying to measure accuracy in predicting both how much the stock price changed and also how close the predicted price was
[2022-11-20 23:42:23] <Lucifer_arma> I'm working on a different metric to determine how accurate each bot was in predicted the direction of the stock price changes
View entire month
DISCLAIMER: These logs of public chat may contain some content which may not be appropriate for all audiences. Use at your own risk.
Logs from 2006-2009 pulled from wrtlprnft
Format changes at: 2015-08-25, 2017-02-20, and 2020-03-23. Times (2015 and later) should be Eastern.