(env, state_winner_triples)
| 337 | |
| 338 | |
| 339 | def initialV_x(env, state_winner_triples): |
| 340 | # initialize state values as follows |
| 341 | # if x wins, V(s) = 1 |
| 342 | # if x loses or draw, V(s) = 0 |
| 343 | # otherwise, V(s) = 0.5 |
| 344 | V = np.zeros(env.num_states) |
| 345 | for state, winner, ended in state_winner_triples: |
| 346 | if ended: |
| 347 | if winner == env.x: |
| 348 | v = 1 |
| 349 | else: |
| 350 | v = 0 |
| 351 | else: |
| 352 | v = 0.5 |
| 353 | V[state] = v |
| 354 | return V |
| 355 | |
| 356 | |
| 357 | def initialV_o(env, state_winner_triples): |