Summary: I’m the tic tac toe guy from a couple weeks ago. I’m working on moving my tic-tac-toe board identification problem to be a full game problem. To start I have been trying to use category encoders rather than scaler encoders for the value of each position on the board. This update is causing …

I was working with this more today, so I wanted to post a quick update. I noticed that if I move the reset column to the end of the row that I don’t hit the assert. I didn’t have time to really dig into the code to figure out why I was hitting the assert in the first place; I was just playing around…

I agree and thought I was moving int hat direction by changing my swarm to use category encoders rather than scaler encoders as a first step. Are you thinking that I should abandon the idea of swarming altogether, or would I come back to it once my encoder is complete? Thanks for helping me underst…

Yes, if I were you I would abandon the idea of swarming and build your own encoder. You want something you can send in a row of input data (in whatever format you want) that returns binary array. The array does not have to be sparse, but be sure to read this paper: [image] Encoding…

Can someone please review my .json and swarm data

Engineering

rhyolight January 5, 2017, 7:13pm 3

I still think you should write a TicTacToe Encoder and forget about swarming. We have some good docs on encoders.

Topic		Replies	Views
Trouble using Category encoder C as special flag and with string datatype NuPIC	8	882	June 15, 2016
Can someone help me figure out how to debug my tic-tac-toe attempt? Engineering usage-help	9	2278	December 15, 2016
Int Assertion Error NuPIC question	41	2758	September 21, 2016
How much should I trust swarm results? NuPIC swarming	38	2110	March 10, 2019
Swarming Nupic Data NuPIC question	7	637	January 2, 2019

Can someone please review my .json and swarm data

Related topics