| Age | Commit message (Expand) | Author |
|---|---|---|
| 44 hours | In-lined score reporting | tslil clingman |
| 44 hours | Finally fixed memory order of weights! | tslil clingman |
| 44 hours | Went back up to 1986 weights | tslil |
| 44 hours | Fixed minimax (!), fixed bugs in tak.c | tslil |
| 44 hours | Auto-generating weights | tslil |
| 44 hours | Generate the correct format directly | tslil |
| 44 hours | one output (of course), Pade approximant of logistic for activation | tslil |
| 44 hours | Working version of this script | tslil |
| 44 hours | Also this was wrong | tslil |
| 44 hours | Two inputs to model, stacks and flat counts | tslil |
| 44 hours | Magic | tslil |
| 44 hours | Closing in on something reasonable for roads | tslil |
| 44 hours | On the hunt for a better representation | tslil |
| 44 hours | Experimenting with training nn | tslil |
