From 232013bda72120f69c1780c4af381884ca111ef4 Mon Sep 17 00:00:00 2001 From: tslil clingman Date: Thu, 21 Jan 2021 00:11:01 -0500 Subject: Experimenting --- include/minimax_cnn1986.c | 120 ++++++++++++++++++++++++---------------------- include/minimax_cnn1986.h | 12 +---- src/ctaklm.c | 10 ++-- 3 files changed, 69 insertions(+), 73 deletions(-) diff --git a/include/minimax_cnn1986.c b/include/minimax_cnn1986.c index e89e5a6..38edb33 100644 --- a/include/minimax_cnn1986.c +++ b/include/minimax_cnn1986.c @@ -36,7 +36,7 @@ static union u_f fudge; #define RELU(x) ((x) = ((x)<0)?0:(x)) float -ct1986_evaluate_black_win(void) { +ct1986_evaluate_win(void) { /* ------------------ * * Convolution layer * * ------------------ */ @@ -114,10 +114,16 @@ ct1986_evaluate_black_win(void) { DOXORSHIFT; output += fudge.f; #endif - if (output > 1.0) return 1.0; - else if (output < 0.0) return 0.0; + if (output > 1.0) { + if (ply & 1) return 1.0; + else return -1.0; + } + else if (output < 0.0) { + if (ply & 1) return -1.0; + else return 1.0; + } - return output; + return (ply & 1) ? 2.0*output-1.0 : 1.0-2*output; } // =================================================================== @@ -138,41 +144,41 @@ previous_ply(void) { static float val; static enum WIN_TYPE w; -#define WIN_EVALUATE_OR_RECURSE(store) { \ +#define WIN_EVALUATE_OR_RECURSE(store,reset) { \ w = 0xFF; \ if (ply >= 2*5 - 2) w = check_win(); \ if (w < 0xFF) { \ - /* Somebody won, assign weights accordingly */ \ - if (min == 0) { \ - if (w == WIN_ROAD_BLACK || w == WIN_FLAT_BLACK) \ - val = infty; \ - else val = -infty; \ - } else { \ + /* Somebody won, assign weights accordingly. Note in particular + that draws are only worth ∞/2 ;) \ + */ \ + if (ply & 1) { \ if (w == WIN_ROAD_WHITE || w == WIN_FLAT_WHITE) \ val = -infty; \ + else if (w == WIN_DRAW) val = infty/2.0; \ + else val = infty; \ + } else { \ + if (w == WIN_ROAD_BLACK || w == WIN_FLAT_BLACK) \ + val = -infty; \ + else if (w == WIN_DRAW) val = infty/2.0; \ else val = infty; \ } \ } else if (cur_depth == max_depth) { \ /* We're at the bottom, evaluate */ \ - val = ct1986_evaluate_black_win(); \ - if ((ply & 1) == 0) val = val - 1.0; \ + val = ct1986_evaluate_win(); \ } else { \ /* We're not at the bottom, recurse first */ \ next_ply(); \ - val = ct1986_minimax(cur_depth + 1, max_depth, 1-min, alpha, beta); \ + val = -ct1986_negamax(cur_depth + 1, max_depth, -beta, -alpha); \ previous_ply(); \ } \ - /* Update the optimal value */ \ - if (((min > 0) && (val <= optimal)) \ - || ((min == 0) && (val >= optimal))) { \ + { reset }; \ + /* Prune */ \ + if (val >= beta) return val; \ + /* Update the optimal value, which alpha carries */ \ + if (val > optimal) { \ optimal = val; \ - if (cur_depth == 0) (store); \ - } \ - /* Update alpha and beta */ \ - if (min) { \ - if (optimal < beta) beta = optimal; \ - } else { \ - if (optimal > alpha) alpha = optimal; \ + if (val > alpha) alpha = val; \ + if (cur_depth == 0) { store }; \ } \ } @@ -180,16 +186,15 @@ static enum WIN_TYPE w; static const int8_t deltas[4] = { +5, -5, -1, +1}; float -ct1986_minimax(const uint8_t cur_depth, const uint8_t max_depth, - const uint8_t min, float alpha, float beta) { +ct1986_negamax(const uint8_t cur_depth, const uint8_t max_depth, + float alpha, float beta) { const uint8_t black = (ply & 1), material = (black) ? black_count : white_count, flat = material & 127, cap = (ply > 2 && (material & 128)), standing = (ply > 2 && (material & 127)); - // 1.0 is a `certain' black win, -1.0 is a `certain' white win. - float optimal = (min) ? infty : -infty; + float optimal = -infty; // Step across the board for (uint8_t row = 0; row < 5; row++) { @@ -297,21 +302,21 @@ ct1986_minimax(const uint8_t cur_depth, const uint8_t max_depth, // If we did update the optimal value, store // this move generate_move(loc, dir, steps, drops, ct1986_ptn); + },{ + // Reset the board data after recursing or + // before returning + if (dir <= M_DOWN) { + for (uint8_t y = 0; y < 5; y++) { + colours[THE_COORDS(col, y)] = colours_backup[y]; + celldat[THE_COORDS(col, y)] = celldat_backup[y]; + } + } else { + for (uint8_t x = 0; x < 5; x++) { + colours[THE_COORDS(x, row)] = colours_backup[x]; + celldat[THE_COORDS(x, row)] = celldat_backup[x]; + } + } }); - // Reset the board data - if (dir <= M_DOWN) { - for (uint8_t y = 0; y < 5; y++) { - colours[THE_COORDS(col, y)] = colours_backup[y]; - celldat[THE_COORDS(col, y)] = celldat_backup[y]; - } - } else { - for (uint8_t x = 0; x < 5; x++) { - colours[THE_COORDS(x, row)] = colours_backup[x]; - celldat[THE_COORDS(x, row)] = celldat_backup[x]; - } - } - // Prune - if (alpha >= beta) return optimal; } /* * With thanks to @@ -337,13 +342,12 @@ ct1986_minimax(const uint8_t cur_depth, const uint8_t max_depth, WIN_EVALUATE_OR_RECURSE({ // If we did update the optimal value, store generate_place(loc, STONE_FLAT, ct1986_ptn); + },{ + // Reset the state + celldat[loc] = 0; + if (black) black_count++; + else white_count++; }); - // Reset the state - celldat[loc] = 0; - if (black) black_count++; - else white_count++; - // Prune - if (alpha >= beta) return optimal; // Do the same for walls, can't happen without flats if (standing) { @@ -353,11 +357,11 @@ ct1986_minimax(const uint8_t cur_depth, const uint8_t max_depth, celldat[loc] = NUM_INC | STONE_STANDING; WIN_EVALUATE_OR_RECURSE({ generate_place(loc, STONE_STANDING, ct1986_ptn); + },{ + celldat[loc] = 0; + if (black) black_count++; + else white_count++; }); - celldat[loc] = 0; - if (black) black_count++; - else white_count++; - if (alpha >= beta) return optimal; } } @@ -369,20 +373,20 @@ ct1986_minimax(const uint8_t cur_depth, const uint8_t max_depth, celldat[loc] = NUM_INC | STONE_CAPSTONE; WIN_EVALUATE_OR_RECURSE({ generate_place(loc, STONE_CAPSTONE, ct1986_ptn); + },{ + celldat[loc] = 0; + if (black) black_count |= 128; + else white_count |= 128; }); - celldat[loc] = 0; - if (black) black_count |= 128; - else white_count |= 128; - if (alpha >= beta) return optimal; } } ct1986_display_progress(cur_depth); } } - return optimal; + return alpha; } inline float ct1986_generate(const uint8_t max_depth) { - return ct1986_minimax(0, 1+2*max_depth, (ply & 1) ? 0 : 1, -infty, infty); + return ct1986_negamax(0, max_depth, -infty, infty); } diff --git a/include/minimax_cnn1986.h b/include/minimax_cnn1986.h index 3ccc004..dd529e6 100644 --- a/include/minimax_cnn1986.h +++ b/include/minimax_cnn1986.h @@ -7,22 +7,14 @@ extern char ct1986_ptn[9]; extern inline void ct1986_display_progress(const uint8_t); float -ct1986_evaluate_black_win(void); +ct1986_evaluate_win(void); // Generate PTN of the ````best'''' action and store it in ct1986_ptn, // along with its value as the return. The ct1986_display_progress // function pointer is called on every new square. -float ct1986_generate(const uint8_t max_depth); - -// Access to the internal method -float -ct1986_minimax(const uint8_t cur_depth, const uint8_t max_depth, - const uint8_t min, float alpha, float beta); - extern char ct1986_ptn[9]; -extern float ct1986_optimal; - +float ct1986_generate(const uint8_t max_depth); void ct1986_generate_ptn(void display_progress(void)); diff --git a/src/ctaklm.c b/src/ctaklm.c index 24613e9..3392af4 100644 --- a/src/ctaklm.c +++ b/src/ctaklm.c @@ -234,7 +234,7 @@ handle_turn(char *line) { static void new_game(uint8_t size) { reset_state(size); - printf("New %dx%d game! ct1986 at search_depth %d.\n", + printf("New %dx%d game! ct1986 at search depth %d.\n", size, size, search_depth); if (gamelog) gamelog = realloc(gamelog, sizeof(char)); else gamelog = malloc(sizeof(char)); @@ -347,13 +347,13 @@ info, load, log, new, play (b|w), self-play, square , ."); } else if (!strcmp(line,"info")) { print_info(); } else if (!strcmp(line,"eval")) { - float eval = ct1986_evaluate_black_win()*100; + float eval = ct1986_evaluate_win()*100; if (ply & 1) { printf("Black heuristic chance: %s%.2f%s\n", blk, eval, rst); } else { printf("White heruistic chance: %s%.2f%s\n", - wht, 100-eval, rst); + wht, eval, rst); } } else if (!strcmp(line,"log")) { puts(gamelog); @@ -404,7 +404,7 @@ info, load, log, new, play (b|w), self-play, square , ."); human = 1 - (line[5] & 1); new_game(5); } else { - puts("Usage: play (b|w).\n"); + puts("Usage: play (b|w)."); } } else { return EXIT_FAILURE; @@ -418,7 +418,7 @@ main(int argc, char **argv) { (void)(argv); human = 0; - search_depth = 1; + search_depth = 3; new_game(5); char *line = NULL; -- cgit v1.2.3 From 9fa3291044ff8d9f0f2b9c9a01cd210bf318ab74 Mon Sep 17 00:00:00 2001 From: tslil clingman Date: Thu, 21 Jan 2021 01:16:20 -0500 Subject: Many fixes, i think this is actually correct --- include/minimax_cnn1986.c | 392 ---------------------------------------------- include/minimax_cnn1986.h | 20 --- include/negamax_cnn1986.c | 389 +++++++++++++++++++++++++++++++++++++++++++++ include/negamax_cnn1986.h | 18 +++ include/tak.c | 2 +- src/ctaklm.c | 25 +-- 6 files changed, 421 insertions(+), 425 deletions(-) delete mode 100644 include/minimax_cnn1986.c delete mode 100644 include/minimax_cnn1986.h create mode 100644 include/negamax_cnn1986.c create mode 100644 include/negamax_cnn1986.h diff --git a/include/minimax_cnn1986.c b/include/minimax_cnn1986.c deleted file mode 100644 index 38edb33..0000000 --- a/include/minimax_cnn1986.c +++ /dev/null @@ -1,392 +0,0 @@ -#include "minimax_cnn1986.h" - -// =================================================================== -// Globals -// =================================================================== - -const float infty = 3.0; -char ct1986_ptn[9]; - -// =================================================================== -// Implementation of a small convolutional neural network -// =================================================================== - -static float flattened[CONV_NUM+2]; -static float dense1[DENSE1_NUM]; -static float dense2[DENSE2_NUM]; - -#ifndef DETERMINISTIC -union u_f { - uint32_t u; - float f; -}; - -static uint32_t state = 1; -static union u_f fudge; - -#define DOXORSHIFT { \ - state ^= state << 13; \ - state ^= state >> 17; \ - state ^= state << 5; \ - fudge.u = 0x3f800000 | state >> 10; \ - fudge.f = (fudge.f - 1.5) * 0.01; \ - } -#endif - -#define RELU(x) ((x) = ((x)<0)?0:(x)) - -float -ct1986_evaluate_win(void) { - /* ------------------ * - * Convolution layer * - * ------------------ */ - // for each kernel - for (uint8_t kern = 0; kern < KERN_NUM; kern++) { - // the stride is 1, march across the board - for (uint8_t bx = 0; bx < KERN_OSIZE; bx++) { - for (uint8_t by = 0; by < KERN_OSIZE; by++) { - flattened[kern+KERN_NUM*(bx+by*KERN_OSIZE)] = - conv2d_biases[kern]; - // Compute the convolution for this position - for (uint8_t ky = 0; ky < KERN_SIZE; ky++) { - for (uint8_t kx = 0; kx < KERN_SIZE; kx++) { - for (uint8_t c = 0; c < KERN_CHAN; c++) { - // Where we are on the board - const uint8_t loc = kx+bx+(ky+by)*5; - // Look up what's on the board at this location, and - // multiply it. For c=0 we have to do some extra work - float lookup = 0; - if (COUNT_AT(loc)>c) { - if (c==0) { - if (STONE_AT(loc) == STONE_STANDING) { - lookup = (colours[loc] & 1) ? +0.25 : -0.25; - } else if (STONE_AT(loc) == STONE_CAPSTONE) { - lookup = (colours[loc] & 1) ? +1.00 : -1.00; - } else { - lookup = (colours[loc] & 1) ? +0.50 : -0.50; - } - } else { - lookup = (colours[loc] & (1< 1.0) { - if (ply & 1) return 1.0; - else return -1.0; - } - else if (output < 0.0) { - if (ply & 1) return -1.0; - else return 1.0; - } - - return (ply & 1) ? 2.0*output-1.0 : 1.0-2*output; -} - -// =================================================================== -// α-β minimax using the above evaluator -// =================================================================== - -static void -previous_ply(void) { - if (ply>0) ply--; - if (ply == 1) { - current_colour = C_WHITE; - } else { - if (current_colour == C_BLACK) current_colour = C_WHITE; - else current_colour = C_BLACK; - } -} - -static float val; -static enum WIN_TYPE w; - -#define WIN_EVALUATE_OR_RECURSE(store,reset) { \ - w = 0xFF; \ - if (ply >= 2*5 - 2) w = check_win(); \ - if (w < 0xFF) { \ - /* Somebody won, assign weights accordingly. Note in particular - that draws are only worth ∞/2 ;) \ - */ \ - if (ply & 1) { \ - if (w == WIN_ROAD_WHITE || w == WIN_FLAT_WHITE) \ - val = -infty; \ - else if (w == WIN_DRAW) val = infty/2.0; \ - else val = infty; \ - } else { \ - if (w == WIN_ROAD_BLACK || w == WIN_FLAT_BLACK) \ - val = -infty; \ - else if (w == WIN_DRAW) val = infty/2.0; \ - else val = infty; \ - } \ - } else if (cur_depth == max_depth) { \ - /* We're at the bottom, evaluate */ \ - val = ct1986_evaluate_win(); \ - } else { \ - /* We're not at the bottom, recurse first */ \ - next_ply(); \ - val = -ct1986_negamax(cur_depth + 1, max_depth, -beta, -alpha); \ - previous_ply(); \ - } \ - { reset }; \ - /* Prune */ \ - if (val >= beta) return val; \ - /* Update the optimal value, which alpha carries */ \ - if (val > optimal) { \ - optimal = val; \ - if (val > alpha) alpha = val; \ - if (cur_depth == 0) { store }; \ - } \ - } - -// UP DOWN LEFT RIGHT -static const int8_t deltas[4] = { +5, -5, -1, +1}; - -float -ct1986_negamax(const uint8_t cur_depth, const uint8_t max_depth, - float alpha, float beta) { - const uint8_t black = (ply & 1), - material = (black) ? black_count : white_count, - flat = material & 127, - cap = (ply > 2 && (material & 128)), - standing = (ply > 2 && (material & 127)); - - float optimal = -infty; - - // Step across the board - for (uint8_t row = 0; row < 5; row++) { - for (uint8_t col = 0; col < 5; col++) { - // Try all valid actions for this square. Is it empty? - const uint8_t loc = THE_COORDS(col, row); - const uint8_t count = (COUNT_AT(loc) > 5) ? 5 : COUNT_AT(loc); - // Only try moves after CPS - if (count && ((colours[loc] & 1) == current_colour) && ply>2) { - // There are stones, can we move them in a given direction? - - // Pre-compute end-stops - uint8_t end_stops[4][2]; // (end, not_crush) - // UP DOWN LEFT RIGHT - end_stops[0][0] = (4-row > count) ? count : 4-row; - end_stops[1][0] = (row > count) ? count : row; - end_stops[2][0] = (col > count) ? count : col; - end_stops[3][0] = (4-col > count) ? count : 4-col; - const uint8_t cap_top = STONE_AT(loc) == STONE_CAPSTONE; - for (uint8_t d = 0; d < 4; d++){ - end_stops[d][1] = 1; - const uint8_t stop = end_stops[d][0]; - end_stops[d][0] = 0; - for (uint8_t k = 1; k <= stop; k++) { - const uint8_t stone = STONE_AT(loc+k*deltas[d]); - if (stone == STONE_STANDING) { - if (cap_top) { - end_stops[d][1] = 0; - end_stops[d][0]++; - } - break; - } else if (stone == STONE_CAPSTONE) { - break; - } - end_stops[d][0]++; - } - } - - uint16_t colours_backup[5]; - uint8_t celldat_backup[5], drops[5]; // we only use 4, the - // fifth is to skip a - // bounds check at (*) - // Back up the row of the board - for (uint8_t y = 0; y < 5; y++) { - colours_backup[y] = colours[THE_COORDS(col, y)]; - celldat_backup[y] = celldat[THE_COORDS(col, y)]; - } - - // I'm not a huge fan of looping through enums, but it's - // better than manually unrolling this. Sufficiently smart - // compilers? - for (enum MOVE_DIRECTION dir = M_UP; dir <= M_RIGHT; dir++) { - // Back-up the column once we start looking horizontally - if (dir == M_LEFT) { - for (uint8_t x = 0; x < 5; x++) { - colours_backup[x] = colours[THE_COORDS(x, row)]; - celldat_backup[x] = celldat[THE_COORDS(x, row)]; - } - } - /* - * We don't do anything terribly efficient here just try - * all the ordered partitions of num ∈ {1 … end_stop}, and - * skip the partition if it calls for multiple stones at - * the end with a crush. - */ - uint8_t gaps, t, idx, mask; - for (uint8_t num = 1; num <= count; num++) { - for (uint8_t steps = 1; - steps <= end_stops[dir][0] && steps <= num; - steps++) { - gaps = 0b00000111 >> (4-steps); - do { - // Translate to a drop sequence - drops[0] = 1; mask = 1; idx = 0; - for (uint8_t d = 0; d + 1 < num; d++) { - if (gaps & mask) { - idx++; - drops[idx] = 1; // (*) we don't need to bounds check - } else { - drops[idx] += 1; - } - mask <<= 1; - } - // TODO: Work out what this should be before partition - - // Ensure legal move if we have to crush - if (end_stops[dir][1] || drops[steps-1] <= 1) { - // Try it, and manually check for win if it's valid - uint8_t j = num; - for (uint8_t k = 0; k < steps; k++) { - // Dear future me, i'm sorry - j -= drops[k]; - colours[loc+(k+1)*deltas[dir]] = (colours[loc+(k+1)*deltas[dir]] << drops[k]) - | ((colours[loc] >> j) & (0xFFFF >> (0x10 - drops[k]))); - celldat[loc+(k+1)*deltas[dir]] = (k == steps - 1) ? STONE_AT(loc) : STONE_FLAT - | ((celldat[loc+(k+1)*deltas[dir]] + ((drops[k] << NUM_SHIFT))) & NUM_MASK); - } - // Then we drop them from the source - colours[loc] >>= num; - const uint8_t dec_count = celldat[loc] - (num << NUM_SHIFT); - celldat[loc] = dec_count & NUM_MASK; - // First check for wins, if we're at the bottom - // evaluate, otherwise recurse - WIN_EVALUATE_OR_RECURSE({ - // If we did update the optimal value, store - // this move - generate_move(loc, dir, steps, drops, ct1986_ptn); - },{ - // Reset the board data after recursing or - // before returning - if (dir <= M_DOWN) { - for (uint8_t y = 0; y < 5; y++) { - colours[THE_COORDS(col, y)] = colours_backup[y]; - celldat[THE_COORDS(col, y)] = celldat_backup[y]; - } - } else { - for (uint8_t x = 0; x < 5; x++) { - colours[THE_COORDS(x, row)] = colours_backup[x]; - celldat[THE_COORDS(x, row)] = celldat_backup[x]; - } - } - }); - } - /* - * With thanks to - * https://graphics.stanford.edu/~seander/bithacks.html#NextBitPermutation - * we have the following magic to generate the next - * permutation of steps-many set bits - */ - t = (gaps | (gaps - 1)); - gaps = (t + 1) | (((~t & -~t) - 1) >> (__builtin_ctz(gaps) + 1)); - } while (gaps && (gaps + 1 <= (1<<(num-1)))); - } - } - } - } else if (material && count == 0) { - // Empty square, try placements - - if (flat) { - // Generate the placement - if (black) black_count--; - else white_count--; - colours[loc] = current_colour; - celldat[loc] = NUM_INC | STONE_FLAT; - WIN_EVALUATE_OR_RECURSE({ - // If we did update the optimal value, store - generate_place(loc, STONE_FLAT, ct1986_ptn); - },{ - // Reset the state - celldat[loc] = 0; - if (black) black_count++; - else white_count++; - }); - - // Do the same for walls, can't happen without flats - if (standing) { - if (black) black_count--; - else white_count--; - colours[loc] = current_colour; - celldat[loc] = NUM_INC | STONE_STANDING; - WIN_EVALUATE_OR_RECURSE({ - generate_place(loc, STONE_STANDING, ct1986_ptn); - },{ - celldat[loc] = 0; - if (black) black_count++; - else white_count++; - }); - } - } - - // and for caps - if (cap) { - if (black) black_count &= 127; - else white_count &= 127; - colours[loc] = current_colour; - celldat[loc] = NUM_INC | STONE_CAPSTONE; - WIN_EVALUATE_OR_RECURSE({ - generate_place(loc, STONE_CAPSTONE, ct1986_ptn); - },{ - celldat[loc] = 0; - if (black) black_count |= 128; - else white_count |= 128; - }); - } - } - ct1986_display_progress(cur_depth); - } - } - return alpha; -} - -inline float -ct1986_generate(const uint8_t max_depth) { - return ct1986_negamax(0, max_depth, -infty, infty); -} diff --git a/include/minimax_cnn1986.h b/include/minimax_cnn1986.h deleted file mode 100644 index dd529e6..0000000 --- a/include/minimax_cnn1986.h +++ /dev/null @@ -1,20 +0,0 @@ -#include -#include "tak.h" -#include "weights.h" - -extern const float infty; -extern char ct1986_ptn[9]; -extern inline void ct1986_display_progress(const uint8_t); - -float -ct1986_evaluate_win(void); - -// Generate PTN of the ````best'''' action and store it in ct1986_ptn, -// along with its value as the return. The ct1986_display_progress -// function pointer is called on every new square. - -extern char ct1986_ptn[9]; -float ct1986_generate(const uint8_t max_depth); - -void -ct1986_generate_ptn(void display_progress(void)); diff --git a/include/negamax_cnn1986.c b/include/negamax_cnn1986.c new file mode 100644 index 0000000..fafb38c --- /dev/null +++ b/include/negamax_cnn1986.c @@ -0,0 +1,389 @@ +#include "negamax_cnn1986.h" + +// =================================================================== +// Globals +// =================================================================== + +const float infty = 3.0; +char ct1986_ptn[9]; +uint8_t ct1986_search_depth = 3; + +// =================================================================== +// Implementation of a small convolutional neural network +// =================================================================== + +static float flattened[CONV_NUM+2]; +static float dense1[DENSE1_NUM]; +static float dense2[DENSE2_NUM]; + +#ifndef DETERMINISTIC +union u_f { + uint32_t u; + float f; +}; + +static uint32_t state = 1; +static union u_f fudge; + +#define DOXORSHIFT { \ + state ^= state << 13; \ + state ^= state >> 17; \ + state ^= state << 5; \ + fudge.u = 0x3f800000 | state >> 10; \ + fudge.f = (fudge.f - 1.5) * 0.01; \ + } +#endif + +#define RELU(x) ((x) = ((x)<0)?0:(x)) + +float +ct1986_evaluate_black_win(void) { + /* ------------------ * + * Convolution layer * + * ------------------ */ + // for each kernel + for (uint8_t kern = 0; kern < KERN_NUM; kern++) { + // the stride is 1, march across the board + for (uint8_t bx = 0; bx < KERN_OSIZE; bx++) { + for (uint8_t by = 0; by < KERN_OSIZE; by++) { + flattened[kern+KERN_NUM*(bx+by*KERN_OSIZE)] = + conv2d_biases[kern]; + // Compute the convolution for this position + for (uint8_t ky = 0; ky < KERN_SIZE; ky++) { + for (uint8_t kx = 0; kx < KERN_SIZE; kx++) { + for (uint8_t c = 0; c < KERN_CHAN; c++) { + // Where we are on the board + const uint8_t loc = kx+bx+(ky+by)*5; + // Look up what's on the board at this location, and + // multiply it. For c=0 we have to do some extra work + float lookup = 0; + if (COUNT_AT(loc)>c) { + if (c==0) { + if (STONE_AT(loc) == STONE_STANDING) { + lookup = (colours[loc] & 1) ? +0.25 : -0.25; + } else if (STONE_AT(loc) == STONE_CAPSTONE) { + lookup = (colours[loc] & 1) ? +1.00 : -1.00; + } else { + lookup = (colours[loc] & 1) ? +0.50 : -0.50; + } + } else { + lookup = (colours[loc] & (1< 1.0) { + return 1.0; + } + else if (output < 0.0) { + return -1.0; + } + return 2*output-1.0; +} + +// =================================================================== +// α-β negamax using the above evaluator +// =================================================================== + +static void +previous_ply(void) { + if (ply>0) ply--; + if (ply == 1) { + current_colour = C_WHITE; + } else { + if (current_colour == C_BLACK) current_colour = C_WHITE; + else current_colour = C_BLACK; + } +} + +static inline void +push_stones(const int8_t location, const uint8_t count, + const uint8_t new_colours, + const enum STONE_VARIANT top_stone) { + colours[location] = (colours[location] << count) | new_colours; + celldat[location] = top_stone + | ((celldat[location] + ((count << NUM_SHIFT))) & NUM_MASK); +} + +static float val; +static enum WIN_TYPE w; + +// UP DOWN LEFT RIGHT +static const int8_t deltas[4] = { +5, -5, -1, +1}; + +#define WIN_EVALUATE_OR_RECURSE(store,reset) { \ + w = 0xFF; \ + if (ply >= 2*5 - 2) w = check_win(); \ + if (w < 0xFF) { \ + /* Somebody won, assign weights accordingly. Note in particular + that draws are only worth ∞/2 ;) + */ \ + if (w == WIN_ROAD_BLACK || w == WIN_FLAT_BLACK) \ + val = infty; \ + else if (w == WIN_DRAW) val = infty/2.0; \ + else val = -infty; \ + val *= colour; \ + } else if (cur_depth == ct1986_search_depth) { \ + /* We're at the bottom, evaluate */ \ + val = colour * ct1986_evaluate_black_win(); \ + } else { \ + /* We're not at the bottom, recurse first */ \ + next_ply(); \ + val = -ct1986_negamax(cur_depth + 1, -beta, -alpha, -colour); \ + previous_ply(); \ + } \ + { reset }; \ + /* Prune */ \ + if (val >= beta) return beta; \ + /* Update the optimal value, which alpha carries */ \ + if (val > alpha) { \ + alpha = val; \ + if (cur_depth == 0) { store }; \ + } \ + } + +float +ct1986_negamax(const uint8_t cur_depth, float alpha, float beta, + const float colour) { + const uint8_t black = (ply & 1), + material = (black) ? black_count : white_count, + flat = material & 127, + cap = (ply > 2 && (material & 128)), + standing = (ply > 2 && (material & 127)); + + // Step across the board + for (uint8_t row = 0; row < 5; row++) { + for (uint8_t col = 0; col < 5; col++) { + // Try all valid actions for this square. Is it empty? + const uint8_t loc = THE_COORDS(col, row); + const uint8_t count = (COUNT_AT(loc) > 5) ? 5 : COUNT_AT(loc); + // Only try moves after CPS + if (count && ((colours[loc] & 1) == current_colour) && ply>2) { + // There are stones, can we move them in a given direction? + + // Pre-compute end-stops + uint8_t end_stops[4][2]; // (end, not_crush) + // UP DOWN LEFT RIGHT + end_stops[0][0] = (4-row > count) ? count : 4-row; + end_stops[1][0] = (row > count) ? count : row; + end_stops[2][0] = (col > count) ? count : col; + end_stops[3][0] = (4-col > count) ? count : 4-col; + const uint8_t cap_top = STONE_AT(loc) == STONE_CAPSTONE; + for (uint8_t d = 0; d < 4; d++){ + end_stops[d][1] = 1; + const uint8_t stop = end_stops[d][0]; + end_stops[d][0] = 0; + for (uint8_t k = 1; k <= stop; k++) { + const uint8_t stone = STONE_AT(loc+k*deltas[d]); + if (stone == STONE_STANDING) { + if (cap_top) { + end_stops[d][1] = 0; + end_stops[d][0]++; + } + break; + } else if (stone == STONE_CAPSTONE) { + break; + } + end_stops[d][0]++; + } + } + + uint16_t colours_backup[5]; + uint8_t celldat_backup[5], drops[5]; // we only use 4, the + // fifth is to skip a + // bounds check at (*) + // Back up the row of the board + for (uint8_t y = 0; y < 5; y++) { + colours_backup[y] = colours[THE_COORDS(col, y)]; + celldat_backup[y] = celldat[THE_COORDS(col, y)]; + } + + // I'm not a huge fan of looping through enums, but it's + // better than manually unrolling this. Sufficiently smart + // compilers? + for (enum MOVE_DIRECTION dir = M_UP; dir <= M_RIGHT; dir++) { + // Back-up the column once we start looking horizontally + if (dir == M_LEFT) { + for (uint8_t x = 0; x < 5; x++) { + colours_backup[x] = colours[THE_COORDS(x, row)]; + celldat_backup[x] = celldat[THE_COORDS(x, row)]; + } + } + /* + * We don't do anything terribly efficient here just try + * all the ordered partitions of num ∈ {1 … end_stop}, and + * skip the partition if it calls for multiple stones at + * the end with a crush. + */ + uint8_t gaps, t, idx, mask; + for (uint8_t num = 1; num <= count; num++) { + for (uint8_t steps = 1; + steps <= end_stops[dir][0] && steps <= num; + steps++) { + gaps = 0b00000111 >> (4-steps); + do { + // Ensure legal move if we have to crush + const uint8_t last_drop_check = (num > 1) ? (gaps & 1<<(num - 2)) : 1; + if (end_stops[dir][1] || last_drop_check) { + // Translate to a drop sequence + drops[0] = 1; mask = 1; idx = 0; + for (uint8_t d = 0; d + 1 < num; d++) { + if (gaps & mask) { + idx++; + drops[idx] = 1; // (*) we don't need to bounds check + } else { + drops[idx] += 1; + } + mask <<= 1; + } + // Do it, and manually check for win if it's valid + uint8_t j = num; + for (uint8_t k = 0; k < steps; k++) { + j -= drops[k]; + push_stones(loc+(k+1)*deltas[dir], + drops[k], + (colours[loc] >> j) & (0xFFFF >> (0x10 - drops[k])), + (k == steps - 1) ? STONE_AT(loc) : STONE_FLAT); + } + // Then we drop them from the source + colours[loc] >>= num; + const uint8_t dec_count = celldat[loc] - (num << NUM_SHIFT); + celldat[loc] = dec_count & NUM_MASK; + + // First check for wins, if we're at the bottom + // evaluate, otherwise recurse + WIN_EVALUATE_OR_RECURSE({ + // If we did update the optimal value, store + // this move + generate_move(loc, dir, steps, drops, ct1986_ptn); + },{ + // Reset the board data after recursing or + // before returning + if (dir <= M_DOWN) { + for (uint8_t y = 0; y < 5; y++) { + colours[THE_COORDS(col, y)] = colours_backup[y]; + celldat[THE_COORDS(col, y)] = celldat_backup[y]; + } + } else { + for (uint8_t x = 0; x < 5; x++) { + colours[THE_COORDS(x, row)] = colours_backup[x]; + celldat[THE_COORDS(x, row)] = celldat_backup[x]; + } + } + }); + } + /* + * With thanks to + * https://graphics.stanford.edu/~seander/bithacks.html#NextBitPermutation + * we have the following magic to generate the next + * permutation of steps-many set bits + */ + t = (gaps | (gaps - 1)); + gaps = (t + 1) | (((~t & -~t) - 1) >> (__builtin_ctz(gaps) + 1)); + } while (gaps && (gaps + 1 <= (1<<(num-1)))); + } + } + } + } else if (material && count == 0) { + // Empty square, try placements + + if (flat) { + // Generate the placement + if (black) black_count--; + else white_count--; + colours[loc] = current_colour; + celldat[loc] = NUM_INC | STONE_FLAT; + WIN_EVALUATE_OR_RECURSE({ + // If we did update the optimal value, store + generate_place(loc, STONE_FLAT, ct1986_ptn); + },{ + // Reset the state + celldat[loc] = 0; + if (black) black_count++; + else white_count++; + }); + + // Do the same for walls, can't happen without flats + if (standing) { + if (black) black_count--; + else white_count--; + colours[loc] = current_colour; + celldat[loc] = NUM_INC | STONE_STANDING; + WIN_EVALUATE_OR_RECURSE({ + generate_place(loc, STONE_STANDING, ct1986_ptn); + },{ + celldat[loc] = 0; + if (black) black_count++; + else white_count++; + }); + } + } + + // and for caps + if (cap) { + if (black) black_count &= 127; + else white_count &= 127; + colours[loc] = current_colour; + celldat[loc] = NUM_INC | STONE_CAPSTONE; + WIN_EVALUATE_OR_RECURSE({ + generate_place(loc, STONE_CAPSTONE, ct1986_ptn); + },{ + celldat[loc] = 0; + if (black) black_count |= 128; + else white_count |= 128; + }); + } + } + ct1986_display_progress(cur_depth); + } + } + return alpha; +} + +inline float +ct1986_generate(void) { + return ct1986_negamax(0, -infty, infty, (ply&1)?1.0:-1.0); +} diff --git a/include/negamax_cnn1986.h b/include/negamax_cnn1986.h new file mode 100644 index 0000000..411d877 --- /dev/null +++ b/include/negamax_cnn1986.h @@ -0,0 +1,18 @@ +#include +#include "tak.h" +#include "weights.h" + +extern const float infty; +extern char ct1986_ptn[9]; +extern uint8_t ct1986_search_depth; +extern inline void ct1986_display_progress(const uint8_t); + +// Do negamax to depth ct1986_search_depth and return PTN of best move +// in ct1986_ptn, along with its value as the return. The +// ct1986_display_progress function is called on every new square at +// the top level. +float ct1986_generate(void); + +// Internal utility function +float +ct1986_evaluate_black_win(void); diff --git a/include/tak.c b/include/tak.c index a748555..e400578 100644 --- a/include/tak.c +++ b/include/tak.c @@ -109,7 +109,7 @@ try_place(const int8_t location, const enum COLOUR colour, // Moving stacks // =================================================================== -static void +static inline void push_stones(const int8_t location, const uint8_t count, const uint8_t new_colours, const enum STONE_VARIANT top_stone) { diff --git a/src/ctaklm.c b/src/ctaklm.c index 3392af4..a1ca69d 100644 --- a/src/ctaklm.c +++ b/src/ctaklm.c @@ -3,7 +3,7 @@ #include #include -#include +#include static const char *blk = "\033[41m", *wht = "\033[44m"; static const char *und = "\033[4m", *rst = "\033[0m"; @@ -151,9 +151,9 @@ print_info(void) { blk, black_count & 127, black_count >> 7, rst); } +static int human; static char *gamelog = 0; static uint8_t auto_board = 0xFF, auto_info = 0xFF; -static int human, search_depth; static void append_to_gamelog(const char *line, const uint8_t win_line) { @@ -235,7 +235,7 @@ static void new_game(uint8_t size) { reset_state(size); printf("New %dx%d game! ct1986 at search depth %d.\n", - size, size, search_depth); + size, size, ct1986_search_depth); if (gamelog) gamelog = realloc(gamelog, sizeof(char)); else gamelog = malloc(sizeof(char)); gamelog[0] = 0; @@ -308,7 +308,7 @@ ct1986_display_progress(const uint8_t depth) { } static int -ct1986_turn(const uint8_t search_depth) { +ct1986_turn(void) { if (won == 0xFF) { // Prepare progress bar fputs("Computing [", stdout); @@ -317,7 +317,7 @@ ct1986_turn(const uint8_t search_depth) { fflush(stdout); // Run the minimax sum_depth = 0; num_check = 0; - float minimax = ct1986_generate(search_depth); + float minimax = ct1986_generate(); fputs("\033[1C ", stdout); // Failed to find a move? if (minimax <= -infty) { @@ -347,24 +347,25 @@ info, load, log, new, play (b|w), self-play, square , ."); } else if (!strcmp(line,"info")) { print_info(); } else if (!strcmp(line,"eval")) { - float eval = ct1986_evaluate_win()*100; + float eval = ct1986_evaluate_black_win()*100; if (ply & 1) { printf("Black heuristic chance: %s%.2f%s\n", blk, eval, rst); } else { printf("White heruistic chance: %s%.2f%s\n", - wht, eval, rst); + wht, -eval, rst); } } else if (!strcmp(line,"log")) { puts(gamelog); } else if (!strcmp(line,"new")) { new_game(5); } else if (!strcmp(line,"self-play")) { - while (ct1986_turn(search_depth) == 0); + while (ct1986_turn() == 0); } else if (!strncmp(line,"depth",5)) { if (strnlen(line,7) == 7 && line[6] >= '0' && line[6] <= '9') { - search_depth = line[6] - '0'; - printf("New search depth: %d.\n",search_depth); + ct1986_search_depth = line[6] - '0'; + printf("New search depth: %d.\n", + ct1986_search_depth); } else { puts("Usage: depth [0-9]."); } @@ -418,7 +419,7 @@ main(int argc, char **argv) { (void)(argv); human = 0; - search_depth = 3; + ct1986_search_depth = 3; new_game(5); char *line = NULL; @@ -446,7 +447,7 @@ main(int argc, char **argv) { if (read == -1) { playing = 0; } else { - ct1986_turn(search_depth); + ct1986_turn(); human = 0; } } -- cgit v1.2.3