├── Chess_AI.ipynb ├── README.md ├── model.h5 ├── modelversion1.h5 ├── modelversion2.h5 ├── modelversion3.h5 └── stockfish └── 13 ├── AUTHORS ├── Copying.txt ├── INSTALL_RECEIPT.json ├── README.md └── bin └── stockfish /Chess_AI.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "code", 5 | "execution_count": 13, 6 | "metadata": { 7 | "id": "QauvWk2MkddY" 8 | }, 9 | "outputs": [], 10 | "source": [ 11 | "import chess\n", 12 | "import chess.engine\n", 13 | "import random\n", 14 | "import numpy\n", 15 | "\n", 16 | "\n", 17 | "# this function will create our x (board)\n", 18 | "def random_board(max_depth=200):\n", 19 | " board = chess.Board()\n", 20 | " depth = random.randrange(0, max_depth)\n", 21 | "\n", 22 | " for _ in range(depth):\n", 23 | " all_moves = list(board.legal_moves)\n", 24 | " random_move = random.choice(all_moves)\n", 25 | " board.push(random_move)\n", 26 | " if board.is_game_over():\n", 27 | " break\n", 28 | "\n", 29 | " return board\n", 30 | "\n", 31 | "\n", 32 | "# this function will create our f(x) (score)\n", 33 | "def stockfish(board, depth):\n", 34 | " with chess.engine.SimpleEngine.popen_uci('stockfish/13/bin/stockfish') as sf:\n", 35 | " result = sf.analyse(board, chess.engine.Limit(depth=depth))\n", 36 | " score = result['score'].white().score()\n", 37 | " return score" 38 | ] 39 | }, 40 | { 41 | "cell_type": "code", 42 | "execution_count": 14, 43 | "metadata": { 44 | "id": "ULOEWyyfYqtq" 45 | }, 46 | "outputs": [ 47 | { 48 | "data": { 49 | "image/svg+xml": [ 50 | "" 51 | ], 52 | "text/plain": [ 53 | "Board('1r6/1qNprk1P/P7/4P1p1/1Rp1b1P1/8/4R3/3K2N1 w - - 1 59')" 54 | ] 55 | }, 56 | "execution_count": 14, 57 | "metadata": {}, 58 | "output_type": "execute_result" 59 | } 60 | ], 61 | "source": [ 62 | "board = random_board()\n", 63 | "board" 64 | ] 65 | }, 66 | { 67 | "cell_type": "code", 68 | "execution_count": 15, 69 | "metadata": { 70 | "id": "QtZy4cR8ZMhq" 71 | }, 72 | "outputs": [ 73 | { 74 | "name": "stdout", 75 | "output_type": "stream", 76 | "text": [ 77 | "769\n" 78 | ] 79 | } 80 | ], 81 | "source": [ 82 | "print(stockfish(board, 10))" 83 | ] 84 | }, 85 | { 86 | "cell_type": "markdown", 87 | "metadata": { 88 | "id": "6Fgr5NTWqgsA" 89 | }, 90 | "source": [ 91 | "The Idea is that the Deep Learning Network will predict this score based on the positions of the pieces on the board." 92 | ] 93 | }, 94 | { 95 | "cell_type": "markdown", 96 | "metadata": { 97 | "id": "cMgBRL1Kdc0B" 98 | }, 99 | "source": [ 100 | "# Creating the dataset" 101 | ] 102 | }, 103 | { 104 | "cell_type": "markdown", 105 | "metadata": { 106 | "id": "d4Sg-yu0sS4t" 107 | }, 108 | "source": [ 109 | "Now we need to convert the board representation to something meaningful.\n", 110 | "A 3d matrix of sizes **8 x 8 x 14** where 8x8 repersents the board and the 14 represents the 7 different pieces " 111 | ] 112 | }, 113 | { 114 | "cell_type": "code", 115 | "execution_count": 16, 116 | "metadata": { 117 | "id": "Rdo64dA7dhBE" 118 | }, 119 | "outputs": [], 120 | "source": [ 121 | "squares_index = {\n", 122 | " 'a': 0,\n", 123 | " 'b': 1,\n", 124 | " 'c': 2,\n", 125 | " 'd': 3,\n", 126 | " 'e': 4,\n", 127 | " 'f': 5,\n", 128 | " 'g': 6,\n", 129 | " 'h': 7\n", 130 | "}\n", 131 | "\n", 132 | "\n", 133 | "# example: h3 -> 17\n", 134 | "def square_to_index(square):\n", 135 | " letter = chess.square_name(square)\n", 136 | " return 8 - int(letter[1]), squares_index[letter[0]]\n", 137 | "\n", 138 | "\n", 139 | "def split_dims(board):\n", 140 | " # this is the 3d matrix\n", 141 | " board3d = numpy.zeros((14, 8, 8), dtype=numpy.int8)\n", 142 | "\n", 143 | " # here we add the pieces's view on the matrix\n", 144 | " for piece in chess.PIECE_TYPES:\n", 145 | " for square in board.pieces(piece, chess.WHITE):\n", 146 | " idx = numpy.unravel_index(square, (8, 8))\n", 147 | " board3d[piece - 1][7 - idx[0]][idx[1]] = 1\n", 148 | " for square in board.pieces(piece, chess.BLACK):\n", 149 | " idx = numpy.unravel_index(square, (8, 8))\n", 150 | " board3d[piece + 5][7 - idx[0]][idx[1]] = 1\n", 151 | "\n", 152 | " # add attacks and valid moves too\n", 153 | " # so the network knows what is being attacked\n", 154 | " aux = board.turn\n", 155 | " board.turn = chess.WHITE\n", 156 | " for move in board.legal_moves:\n", 157 | " i, j = square_to_index(move.to_square)\n", 158 | " board3d[12][i][j] = 1\n", 159 | " board.turn = chess.BLACK\n", 160 | " for move in board.legal_moves:\n", 161 | " i, j = square_to_index(move.to_square)\n", 162 | " board3d[13][i][j] = 1\n", 163 | " board.turn = aux\n", 164 | "\n", 165 | " return board3d" 166 | ] 167 | }, 168 | { 169 | "cell_type": "code", 170 | "execution_count": 17, 171 | "metadata": { 172 | "id": "gHONl_M1hG9i" 173 | }, 174 | "outputs": [ 175 | { 176 | "data": { 177 | "text/plain": [ 178 | "array([[[0, 0, 0, 0, 0, 0, 0, 0],\n", 179 | " [0, 0, 0, 0, 0, 0, 0, 1],\n", 180 | " [1, 0, 0, 0, 0, 0, 0, 0],\n", 181 | " [0, 0, 0, 0, 1, 0, 0, 0],\n", 182 | " [0, 0, 0, 0, 0, 0, 1, 0],\n", 183 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 184 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 185 | " [0, 0, 0, 0, 0, 0, 0, 0]],\n", 186 | "\n", 187 | " [[0, 0, 0, 0, 0, 0, 0, 0],\n", 188 | " [0, 0, 1, 0, 0, 0, 0, 0],\n", 189 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 190 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 191 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 192 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 193 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 194 | " [0, 0, 0, 0, 0, 0, 1, 0]],\n", 195 | "\n", 196 | " [[0, 0, 0, 0, 0, 0, 0, 0],\n", 197 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 198 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 199 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 200 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 201 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 202 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 203 | " [0, 0, 0, 0, 0, 0, 0, 0]],\n", 204 | "\n", 205 | " [[0, 0, 0, 0, 0, 0, 0, 0],\n", 206 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 207 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 208 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 209 | " [0, 1, 0, 0, 0, 0, 0, 0],\n", 210 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 211 | " [0, 0, 0, 0, 1, 0, 0, 0],\n", 212 | " [0, 0, 0, 0, 0, 0, 0, 0]],\n", 213 | "\n", 214 | " [[0, 0, 0, 0, 0, 0, 0, 0],\n", 215 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 216 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 217 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 218 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 219 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 220 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 221 | " [0, 0, 0, 0, 0, 0, 0, 0]],\n", 222 | "\n", 223 | " [[0, 0, 0, 0, 0, 0, 0, 0],\n", 224 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 225 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 226 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 227 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 228 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 229 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 230 | " [0, 0, 0, 1, 0, 0, 0, 0]],\n", 231 | "\n", 232 | " [[0, 0, 0, 0, 0, 0, 0, 0],\n", 233 | " [0, 0, 0, 1, 0, 0, 0, 0],\n", 234 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 235 | " [0, 0, 0, 0, 0, 0, 1, 0],\n", 236 | " [0, 0, 1, 0, 0, 0, 0, 0],\n", 237 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 238 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 239 | " [0, 0, 0, 0, 0, 0, 0, 0]],\n", 240 | "\n", 241 | " [[0, 0, 0, 0, 0, 0, 0, 0],\n", 242 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 243 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 244 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 245 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 246 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 247 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 248 | " [0, 0, 0, 0, 0, 0, 0, 0]],\n", 249 | "\n", 250 | " [[0, 0, 0, 0, 0, 0, 0, 0],\n", 251 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 252 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 253 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 254 | " [0, 0, 0, 0, 1, 0, 0, 0],\n", 255 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 256 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 257 | " [0, 0, 0, 0, 0, 0, 0, 0]],\n", 258 | "\n", 259 | " [[0, 1, 0, 0, 0, 0, 0, 0],\n", 260 | " [0, 0, 0, 0, 1, 0, 0, 0],\n", 261 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 262 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 263 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 264 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 265 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 266 | " [0, 0, 0, 0, 0, 0, 0, 0]],\n", 267 | "\n", 268 | " [[0, 0, 0, 0, 0, 0, 0, 0],\n", 269 | " [0, 1, 0, 0, 0, 0, 0, 0],\n", 270 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 271 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 272 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 273 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 274 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 275 | " [0, 0, 0, 0, 0, 0, 0, 0]],\n", 276 | "\n", 277 | " [[0, 0, 0, 0, 0, 0, 0, 0],\n", 278 | " [0, 0, 0, 0, 0, 1, 0, 0],\n", 279 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 280 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 281 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 282 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 283 | " [0, 0, 0, 0, 0, 0, 0, 0],\n", 284 | " [0, 0, 0, 0, 0, 0, 0, 0]],\n", 285 | "\n", 286 | " [[1, 0, 0, 0, 1, 0, 0, 1],\n", 287 | " [1, 1, 0, 0, 0, 0, 0, 0],\n", 288 | " [0, 1, 0, 0, 1, 0, 0, 0],\n", 289 | " [0, 1, 0, 1, 0, 0, 0, 0],\n", 290 | " [1, 0, 1, 0, 1, 0, 0, 0],\n", 291 | " [0, 1, 0, 0, 1, 1, 0, 1],\n", 292 | " [1, 1, 1, 1, 0, 1, 1, 1],\n", 293 | " [0, 1, 1, 0, 1, 0, 0, 0]],\n", 294 | "\n", 295 | " [[1, 0, 1, 1, 1, 1, 1, 1],\n", 296 | " [1, 0, 1, 0, 0, 0, 1, 1],\n", 297 | " [1, 1, 1, 1, 1, 0, 1, 0],\n", 298 | " [0, 1, 0, 1, 1, 1, 0, 0],\n", 299 | " [0, 1, 0, 0, 0, 0, 0, 0],\n", 300 | " [0, 0, 1, 1, 0, 1, 0, 0],\n", 301 | " [0, 0, 1, 0, 0, 0, 1, 0],\n", 302 | " [0, 1, 0, 0, 0, 0, 0, 1]]], dtype=int8)" 303 | ] 304 | }, 305 | "execution_count": 17, 306 | "metadata": {}, 307 | "output_type": "execute_result" 308 | } 309 | ], 310 | "source": [ 311 | "split_dims(board)" 312 | ] 313 | }, 314 | { 315 | "cell_type": "markdown", 316 | "metadata": { 317 | "id": "OwpzPcsXfuVw" 318 | }, 319 | "source": [ 320 | "Now, all we have to do is call **random_board()** to create random boards, **stockfish()** to get a score for how good each board is for white.\n", 321 | "\n", 322 | "Then we convert each board to a 3d matrix using **split_dims()**, now creating the dataset is easy!" 323 | ] 324 | }, 325 | { 326 | "cell_type": "markdown", 327 | "metadata": { 328 | "id": "ukmA7z-dlB6m" 329 | }, 330 | "source": [ 331 | "# TensorFlow!" 332 | ] 333 | }, 334 | { 335 | "cell_type": "code", 336 | "execution_count": 18, 337 | "metadata": { 338 | "id": "6S7QNZqwmBOP" 339 | }, 340 | "outputs": [], 341 | "source": [ 342 | "import tensorflow.keras.models as models\n", 343 | "import tensorflow.keras.layers as layers\n", 344 | "import tensorflow.keras.utils as utils\n", 345 | "import tensorflow.keras.optimizers as optimizers\n", 346 | "\n", 347 | "\n", 348 | "def build_model(conv_size, conv_depth):\n", 349 | " board3d = layers.Input(shape=(14, 8, 8))\n", 350 | "\n", 351 | " # adding the convolutional layers\n", 352 | " x = board3d\n", 353 | " for _ in range(conv_depth):\n", 354 | " x = layers.Conv2D(filters=conv_size, kernel_size=3, padding='same', activation='relu')(x)\n", 355 | " x = layers.Flatten()(x)\n", 356 | " x = layers.Dense(64, 'relu')(x)\n", 357 | " x = layers.Dense(1, 'sigmoid')(x)\n", 358 | "\n", 359 | " return models.Model(inputs=board3d, outputs=x)" 360 | ] 361 | }, 362 | { 363 | "cell_type": "markdown", 364 | "metadata": { 365 | "id": "6S1RCn3xn4Bu" 366 | }, 367 | "source": [ 368 | "Skip connections (residual network) will likely improve the model for deeper connections. If you want to test the residual model, check the code below." 369 | ] 370 | }, 371 | { 372 | "cell_type": "code", 373 | "execution_count": 19, 374 | "metadata": { 375 | "id": "tAFSOFc8pJf8" 376 | }, 377 | "outputs": [], 378 | "source": [ 379 | "def build_model_residual(conv_size, conv_depth):\n", 380 | " board3d = layers.Input(shape=(14, 8, 8))\n", 381 | "\n", 382 | " # adding the convolutional layers\n", 383 | " x = layers.Conv2D(filters=conv_size, kernel_size=3, padding='same')(board3d)\n", 384 | " for _ in range(conv_depth):\n", 385 | " previous = x\n", 386 | " x = layers.Conv2D(filters=conv_size, kernel_size=3, padding='same')(x)\n", 387 | " x = layers.BatchNormalization()(x)\n", 388 | " x = layers.Activation('relu')(x)\n", 389 | " x = layers.Conv2D(filters=conv_size, kernel_size=3, padding='same')(x)\n", 390 | " x = layers.BatchNormalization()(x)\n", 391 | " x = layers.Add()([x, previous])\n", 392 | " x = layers.Activation('relu')(x)\n", 393 | " x = layers.Flatten()(x)\n", 394 | " x = layers.Dense(1, 'sigmoid')(x)\n", 395 | "\n", 396 | " return models.Model(inputs=board3d, outputs=x)" 397 | ] 398 | }, 399 | { 400 | "cell_type": "code", 401 | "execution_count": 20, 402 | "metadata": { 403 | "id": "3IjDiS3Bmo5m" 404 | }, 405 | "outputs": [], 406 | "source": [ 407 | "model = build_model(32, 4)" 408 | ] 409 | }, 410 | { 411 | "cell_type": "markdown", 412 | "metadata": { 413 | "id": "ck79-w2ZxwVB" 414 | }, 415 | "source": [ 416 | "# It's training time!" 417 | ] 418 | }, 419 | { 420 | "cell_type": "code", 421 | "execution_count": null, 422 | "metadata": { 423 | "cellView": "both", 424 | "id": "CkOXxmoVyHdc" 425 | }, 426 | "outputs": [], 427 | "source": [ 428 | "import tensorflow.keras.callbacks as callbacks\n", 429 | "\n", 430 | "\n", 431 | "def get_dataset():\n", 432 | "\tcontainer = numpy.load('dataset.npz')\n", 433 | "\tb, v = container['b'], container['v']\n", 434 | "\tv = numpy.asarray(v / abs(v).max() / 2 + 0.5, dtype=numpy.float32) # normalization (0 - 1)\n", 435 | "\treturn b, v\n", 436 | "\n", 437 | "\n", 438 | "x_train, y_train = get_dataset()\n", 439 | "x_train.transpose()\n", 440 | "print(x_train.shape)\n", 441 | "print(y_train.shape)" 442 | ] 443 | }, 444 | { 445 | "cell_type": "code", 446 | "execution_count": null, 447 | "metadata": { 448 | "id": "RyOYq9mv2ppC" 449 | }, 450 | "outputs": [], 451 | "source": [ 452 | "from tensorflow.keras.callbacks import ModelCheckpoint\n", 453 | "model.compile(optimizer=optimizers.Adam(5e-4), loss='mean_squared_error')\n", 454 | "model.summary()\n", 455 | "checkpoint_filepath = '/tmp/checkpoint/'\n", 456 | "model_checkpointing_callback = ModelCheckpoint(\n", 457 | " filepath = checkpoint_filepath,\n", 458 | " save_best_only= True,\n", 459 | ")\n", 460 | "model.fit(x_train, y_train,\n", 461 | " batch_size=2048,\n", 462 | " epochs=1000,\n", 463 | " verbose=1,\n", 464 | " validation_split=0.1,\n", 465 | " callbacks=[callbacks.ReduceLROnPlateau(monitor='loss', patience=10),\n", 466 | " callbacks.EarlyStopping(monitor='loss', patience=15, min_delta=1e-4),model_checkpointing_callback])\n", 467 | "\n", 468 | "model.save('model.h5')" 469 | ] 470 | }, 471 | { 472 | "cell_type": "markdown", 473 | "metadata": { 474 | "id": "UndQQeUurAKp" 475 | }, 476 | "source": [ 477 | "# Playing with the AI" 478 | ] 479 | }, 480 | { 481 | "cell_type": "code", 482 | "execution_count": 21, 483 | "metadata": {}, 484 | "outputs": [], 485 | "source": [ 486 | "from tensorflow.keras import models\n", 487 | "model = models.load_model('model.h5')" 488 | ] 489 | }, 490 | { 491 | "cell_type": "code", 492 | "execution_count": 22, 493 | "metadata": { 494 | "id": "e4CfjcGorHzg" 495 | }, 496 | "outputs": [], 497 | "source": [ 498 | "# used for the minimax algorithm\n", 499 | "def minimax_eval(board):\n", 500 | " board3d = split_dims(board)\n", 501 | " board3d = numpy.expand_dims(board3d, 0)\n", 502 | " return model(board3d)[0][0]\n", 503 | "\n", 504 | "\n", 505 | "def minimax(board, depth, alpha, beta, maximizing_player):\n", 506 | " if depth == 0 or board.is_game_over():\n", 507 | " return minimax_eval(board)\n", 508 | " \n", 509 | " if maximizing_player:\n", 510 | " max_eval = -numpy.inf\n", 511 | " for move in board.legal_moves:\n", 512 | " board.push(move)\n", 513 | " eval = minimax(board, depth - 1, alpha, beta, False)\n", 514 | " board.pop()\n", 515 | " max_eval = max(max_eval, eval)\n", 516 | " alpha = max(alpha, eval)\n", 517 | " if beta <= alpha:\n", 518 | " break\n", 519 | " return max_eval\n", 520 | " else:\n", 521 | " min_eval = numpy.inf\n", 522 | " for move in board.legal_moves:\n", 523 | " board.push(move)\n", 524 | " eval = minimax(board, depth - 1, alpha, beta, True)\n", 525 | " board.pop()\n", 526 | " min_eval = min(min_eval, eval)\n", 527 | " beta = min(beta, eval)\n", 528 | " if beta <= alpha:\n", 529 | " break\n", 530 | " return min_eval\n", 531 | "\n", 532 | "\n", 533 | "# this is the actual function that gets the move from the neural network\n", 534 | "def get_ai_move(board, depth):\n", 535 | " max_move = None\n", 536 | " max_eval = -numpy.inf\n", 537 | "\n", 538 | " for move in board.legal_moves:\n", 539 | " board.push(move)\n", 540 | " eval = minimax(board, depth - 1, -numpy.inf, numpy.inf, False)\n", 541 | " board.pop()\n", 542 | " if eval > max_eval:\n", 543 | " max_eval = eval\n", 544 | " max_move = move\n", 545 | " \n", 546 | " return max_move" 547 | ] 548 | }, 549 | { 550 | "cell_type": "code", 551 | "execution_count": 25, 552 | "metadata": {}, 553 | "outputs": [ 554 | { 555 | "name": "stdout", 556 | "output_type": "stream", 557 | "text": [ 558 | "\n", 559 | "r . b . . . k .\n", 560 | "p p . . . . p p\n", 561 | ". . . K . . . .\n", 562 | ". . . . . r . .\n", 563 | ". . . . q . . .\n", 564 | ". . . . . . . .\n", 565 | ". P P . . . P P\n", 566 | ". . . . . . . .\n", 567 | "\n", 568 | "r . b . . . k .\n", 569 | "p p . . . . p p\n", 570 | ". . . K . . . .\n", 571 | ". . . . q r . .\n", 572 | ". . . . . . . .\n", 573 | ". . . . . . . .\n", 574 | ". P P . . . P P\n", 575 | ". . . . . . . .\n", 576 | "game_over\n" 577 | ] 578 | } 579 | ], 580 | "source": [ 581 | "# Testing code AI(white) vs Stockfish(black)\n", 582 | "board = chess.Board()\n", 583 | "from IPython.display import clear_output\n", 584 | "\n", 585 | "with chess.engine.SimpleEngine.popen_uci('stockfish/7/bin/stockfish') as engine:\n", 586 | " while True:\n", 587 | " clear_output(wait=True)\n", 588 | " move = get_ai_move(board, 1)\n", 589 | " board.push(move)\n", 590 | " print(f'\\n{board}')\n", 591 | " if board.is_game_over():\n", 592 | " print('game_over')\n", 593 | " break\n", 594 | " move = engine.analyse(board, chess.engine.Limit(time=0.1), info=chess.engine.INFO_PV)['pv'][0]\n", 595 | " board.push(move)\n", 596 | " print(f'\\n{board}')\n", 597 | " if board.is_game_over():\n", 598 | " print('game_over')\n", 599 | " break" 600 | ] 601 | }, 602 | { 603 | "cell_type": "code", 604 | "execution_count": null, 605 | "metadata": { 606 | "id": "C63ND0E_uffp" 607 | }, 608 | "outputs": [], 609 | "source": [ 610 | "# Move by move testing code AI(white) vs Stockfish(black)\n", 611 | "board = chess.Board()\n", 612 | "\n", 613 | "with chess.engine.SimpleEngine.popen_uci('stockfish/13/bin/stockfish') as engine:\n", 614 | " while True:\n", 615 | " move = get_ai_move(board, 1)\n", 616 | " board.push(move)\n", 617 | " print(f'\\n{board}')\n", 618 | " if board.is_game_over():\n", 619 | " print('game_over')\n", 620 | " break\n", 621 | " move = engine.analyse(board, chess.engine.Limit(time=1), info=chess.engine.INFO_PV)['pv'][0]\n", 622 | " board.push(move)\n", 623 | " print(f'\\n{board}')\n", 624 | " if board.is_game_over():\n", 625 | " print('game_over')\n", 626 | " break" 627 | ] 628 | }, 629 | { 630 | "cell_type": "code", 631 | "execution_count": null, 632 | "metadata": {}, 633 | "outputs": [], 634 | "source": [ 635 | "# Move by move testing code AI(white) vs Stockfish(black)\n", 636 | "board = chess.Board()\n", 637 | "from IPython.display import clear_output\n", 638 | "\n", 639 | "with chess.engine.SimpleEngine.popen_uci('stockfish/13/bin/stockfish') as engine:\n", 640 | " while True:\n", 641 | " clear_output(wait=True)\n", 642 | " move = get_ai_move(board, 1)\n", 643 | " board.push(move)\n", 644 | " print(move)\n", 645 | " print(f'\\n{board}')\n", 646 | " if board.is_game_over():\n", 647 | " print('game_over')\n", 648 | " break\n", 649 | " input_var = input()\n", 650 | " move = chess.Move.from_uci(input_var)\n", 651 | " board.push(move)\n", 652 | " print(move)\n", 653 | " print(f'\\n{board}')\n", 654 | " if board.is_game_over():\n", 655 | " print('game_over')\n", 656 | " break" 657 | ] 658 | }, 659 | { 660 | "cell_type": "code", 661 | "execution_count": null, 662 | "metadata": {}, 663 | "outputs": [], 664 | "source": [] 665 | } 666 | ], 667 | "metadata": { 668 | "accelerator": "GPU", 669 | "colab": { 670 | "collapsed_sections": [], 671 | "name": "Chess AI", 672 | "private_outputs": true, 673 | "provenance": [] 674 | }, 675 | "interpreter": { 676 | "hash": "aee8b7b246df8f9039afb4144a1f6fd8d2ca17a180786b69acc140d282b71a49" 677 | }, 678 | "kernelspec": { 679 | "display_name": "Python 3 (ipykernel)", 680 | "language": "python", 681 | "name": "python3" 682 | }, 683 | "language_info": { 684 | "codemirror_mode": { 685 | "name": "ipython", 686 | "version": 3 687 | }, 688 | "file_extension": ".py", 689 | "mimetype": "text/x-python", 690 | "name": "python", 691 | "nbconvert_exporter": "python", 692 | "pygments_lexer": "ipython3", 693 | "version": "3.8.0" 694 | } 695 | }, 696 | "nbformat": 4, 697 | "nbformat_minor": 1 698 | } 699 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Chess-AI-with-TensorFlow -------------------------------------------------------------------------------- /model.h5: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/realnihal/Chess-AI-with-TensorFlow/2ecda18b164ec63fb38cf97b919845e595b83df0/model.h5 -------------------------------------------------------------------------------- /modelversion1.h5: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/realnihal/Chess-AI-with-TensorFlow/2ecda18b164ec63fb38cf97b919845e595b83df0/modelversion1.h5 -------------------------------------------------------------------------------- /modelversion2.h5: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/realnihal/Chess-AI-with-TensorFlow/2ecda18b164ec63fb38cf97b919845e595b83df0/modelversion2.h5 -------------------------------------------------------------------------------- /modelversion3.h5: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/realnihal/Chess-AI-with-TensorFlow/2ecda18b164ec63fb38cf97b919845e595b83df0/modelversion3.h5 -------------------------------------------------------------------------------- /stockfish/13/AUTHORS: -------------------------------------------------------------------------------- 1 | # List of authors for Stockfish, as of August 4, 2020 2 | 3 | # Founders of the Stockfish project and fishtest infrastructure 4 | Tord Romstad (romstad) 5 | Marco Costalba (mcostalba) 6 | Joona Kiiski (zamar) 7 | Gary Linscott (glinscott) 8 | 9 | # Authors and inventors of NNUE, training, NNUE port 10 | Yu Nasu (ynasu87) 11 | Motohiro Isozaki (yaneurao) 12 | Hisayori Noda (nodchip) 13 | 14 | # all other authors of the code in alphabetical order 15 | Aditya (absimaldata) 16 | Adrian Petrescu (apetresc) 17 | Ajith Chandy Jose (ajithcj) 18 | Alain Savard (Rocky640) 19 | Alayan Feh (Alayan-stk-2) 20 | Alexander Kure 21 | Alexander Pagel (Lolligerhans) 22 | Alfredo Menezes (lonfom169) 23 | Ali AlZhrani (Cooffe) 24 | Andrew Grant (AndyGrant) 25 | Andrey Neporada (nepal) 26 | Andy Duplain 27 | Aram Tumanian (atumanian) 28 | Arjun Temurnikar 29 | Auguste Pop 30 | Balint Pfliegel 31 | Ben Koshy (BKSpurgeon) 32 | Bill Henry (VoyagerOne) 33 | Bojun Guo (noobpwnftw, Nooby) 34 | braich 35 | Brian Sheppard (SapphireBrand, briansheppard-toast) 36 | Bruno de Melo Costa (BM123499) 37 | Bryan Cross (crossbr) 38 | candirufish 39 | Chess13234 40 | Chris Cain (ceebo) 41 | Dale Weiler (graphitemaster) 42 | Dan Schmidt (dfannius) 43 | Daniel Axtens (daxtens) 44 | Daniel Dugovic (ddugovic) 45 | Dariusz Orzechowski (dorzechowski) 46 | David Zar 47 | Daylen Yang (daylen) 48 | Deshawn Mohan-Smith (GoldenRare) 49 | Dieter Dobbelaere (ddobbelaere) 50 | DiscanX 51 | Dominik Schlösser (domschl) 52 | double-beep 53 | Eduardo Cáceres (eduherminio) 54 | Eelco de Groot (KingDefender) 55 | Elvin Liu (solarlight2) 56 | erbsenzaehler 57 | Ernesto Gatti 58 | Linmiao Xu (linrock) 59 | Fabian Beuke (madnight) 60 | Fabian Fichter (ianfab) 61 | Fanael Linithien (Fanael) 62 | fanon 63 | Fauzi Akram Dabat (FauziAkram) 64 | Felix Wittmann 65 | gamander 66 | Gary Heckman (gheckman) 67 | George Sobala (gsobala) 68 | gguliash 69 | Gian-Carlo Pascutto (gcp) 70 | Gontran Lemaire (gonlem) 71 | Goodkov Vasiliy Aleksandrovich (goodkov) 72 | Gregor Cramer 73 | GuardianRM 74 | Günther Demetz (pb00067, pb00068) 75 | Guy Vreuls (gvreuls) 76 | Henri Wiechers 77 | Hiraoka Takuya (HiraokaTakuya) 78 | homoSapiensSapiens 79 | Hongzhi Cheng 80 | Ivan Ivec (IIvec) 81 | Jacques B. (Timshel) 82 | Jan Ondruš (hxim) 83 | Jared Kish (Kurtbusch) 84 | Jarrod Torriero (DU-jdto) 85 | Jean Gauthier (OuaisBla) 86 | Jean-Francois Romang (jromang) 87 | Jekaa 88 | Jerry Donald Watson (jerrydonaldwatson) 89 | jjoshua2 90 | Jonathan Calovski (Mysseno) 91 | Jonathan Buladas Dumale (SFisGOD) 92 | Joost VandeVondele (vondele) 93 | Jörg Oster (joergoster) 94 | Joseph Ellis (jhellis3) 95 | Joseph R. Prostko 96 | jundery 97 | Justin Blanchard (UncombedCoconut) 98 | Kelly Wilson 99 | Ken Takusagawa 100 | kinderchocolate 101 | Kiran Panditrao (Krgp) 102 | Kojirion 103 | Krystian Kuzniarek (kuzkry) 104 | Leonardo Ljubičić (ICCF World Champion) 105 | Leonid Pechenik (lp--) 106 | Linus Arver (listx) 107 | loco-loco 108 | Lub van den Berg (ElbertoOne) 109 | Luca Brivio (lucabrivio) 110 | Lucas Braesch (lucasart) 111 | Lyudmil Antonov (lantonov) 112 | Maciej Żenczykowski (zenczykowski) 113 | Malcolm Campbell (xoto10) 114 | Mark Tenzer (31m059) 115 | marotear 116 | Matt Ginsberg (mattginsberg) 117 | Matthew Lai (matthewlai) 118 | Matthew Sullivan (Matt14916) 119 | Maxim Molchanov (Maxim) 120 | Michael An (man) 121 | Michael Byrne (MichaelB7) 122 | Michael Chaly (Vizvezdenec) 123 | Michael Stembera (mstembera) 124 | Michael Whiteley (protonspring) 125 | Michel Van den Bergh (vdbergh) 126 | Miguel Lahoz (miguel-l) 127 | Mikael Bäckman (mbootsector) 128 | Mira 129 | Miroslav Fontán (Hexik) 130 | Moez Jellouli (MJZ1977) 131 | Mohammed Li (tthsqe12) 132 | Nathan Rugg (nmrugg) 133 | Nick Pelling (nickpelling) 134 | Nicklas Persson (NicklasPersson) 135 | Niklas Fiekas (niklasf) 136 | Nikolay Kostov (NikolayIT) 137 | Nguyen Pham (nguyenpham) 138 | Norman Schmidt (FireFather) 139 | notruck 140 | Ondrej Mosnáček (WOnder93) 141 | Oskar Werkelin Ahlin 142 | Pablo Vazquez 143 | Panthee 144 | Pascal Romaret 145 | Pasquale Pigazzini (ppigazzini) 146 | Patrick Jansen (mibere) 147 | pellanda 148 | Peter Zsifkovits (CoffeeOne) 149 | Praveen Kumar Tummala (praveentml) 150 | Rahul Dsilva (silversolver1) 151 | Ralph Stößer (Ralph Stoesser) 152 | Raminder Singh 153 | renouve 154 | Reuven Peleg 155 | Richard Lloyd 156 | Rodrigo Exterckötter Tjäder 157 | Ron Britvich (Britvich) 158 | Ronald de Man (syzygy1, syzygy) 159 | rqs 160 | Ryan Schmitt 161 | Ryan Takker 162 | Sami Kiminki (skiminki) 163 | Sebastian Buchwald (UniQP) 164 | Sergei Antonov (saproj) 165 | Sergei Ivanov (svivanov72) 166 | Sergio Vieri (sergiovieri) 167 | sf-x 168 | Shane Booth (shane31) 169 | Shawn Varghese (xXH4CKST3RXx) 170 | Stefan Geschwentner (locutus2) 171 | Stefano Cardanobile (Stefano80) 172 | Steinar Gunderson (sesse) 173 | Stéphane Nicolet (snicolet) 174 | Thanar2 175 | thaspel 176 | theo77186 177 | Tom Truscott 178 | Tom Vijlbrief (tomtor) 179 | Tomasz Sobczyk (Sopel97) 180 | Torsten Franz (torfranz, tfranzer) 181 | Tracey Emery (basepr1me) 182 | tttak 183 | Unai Corzo (unaiic) 184 | Uri Blass (uriblass) 185 | Vince Negri (cuddlestmonkey) 186 | zz4032 187 | 188 | 189 | # Additionally, we acknowledge the authors and maintainers of fishtest, 190 | # an amazing and essential framework for the development of Stockfish! 191 | # 192 | # https://github.com/glinscott/fishtest/blob/master/AUTHORS 193 | -------------------------------------------------------------------------------- /stockfish/13/Copying.txt: -------------------------------------------------------------------------------- 1 | GNU GENERAL PUBLIC LICENSE 2 | Version 3, 29 June 2007 3 | 4 | Copyright (C) 2007 Free Software Foundation, Inc. 5 | Everyone is permitted to copy and distribute verbatim copies 6 | of this license document, but changing it is not allowed. 7 | 8 | Preamble 9 | 10 | The GNU General Public License is a free, copyleft license for 11 | software and other kinds of works. 12 | 13 | The licenses for most software and other practical works are designed 14 | to take away your freedom to share and change the works. By contrast, 15 | the GNU General Public License is intended to guarantee your freedom to 16 | share and change all versions of a program--to make sure it remains free 17 | software for all its users. We, the Free Software Foundation, use the 18 | GNU General Public License for most of our software; it applies also to 19 | any other work released this way by its authors. You can apply it to 20 | your programs, too. 21 | 22 | When we speak of free software, we are referring to freedom, not 23 | price. Our General Public Licenses are designed to make sure that you 24 | have the freedom to distribute copies of free software (and charge for 25 | them if you wish), that you receive source code or can get it if you 26 | want it, that you can change the software or use pieces of it in new 27 | free programs, and that you know you can do these things. 28 | 29 | To protect your rights, we need to prevent others from denying you 30 | these rights or asking you to surrender the rights. Therefore, you have 31 | certain responsibilities if you distribute copies of the software, or if 32 | you modify it: responsibilities to respect the freedom of others. 33 | 34 | For example, if you distribute copies of such a program, whether 35 | gratis or for a fee, you must pass on to the recipients the same 36 | freedoms that you received. You must make sure that they, too, receive 37 | or can get the source code. And you must show them these terms so they 38 | know their rights. 39 | 40 | Developers that use the GNU GPL protect your rights with two steps: 41 | (1) assert copyright on the software, and (2) offer you this License 42 | giving you legal permission to copy, distribute and/or modify it. 43 | 44 | For the developers' and authors' protection, the GPL clearly explains 45 | that there is no warranty for this free software. For both users' and 46 | authors' sake, the GPL requires that modified versions be marked as 47 | changed, so that their problems will not be attributed erroneously to 48 | authors of previous versions. 49 | 50 | Some devices are designed to deny users access to install or run 51 | modified versions of the software inside them, although the manufacturer 52 | can do so. This is fundamentally incompatible with the aim of 53 | protecting users' freedom to change the software. The systematic 54 | pattern of such abuse occurs in the area of products for individuals to 55 | use, which is precisely where it is most unacceptable. Therefore, we 56 | have designed this version of the GPL to prohibit the practice for those 57 | products. If such problems arise substantially in other domains, we 58 | stand ready to extend this provision to those domains in future versions 59 | of the GPL, as needed to protect the freedom of users. 60 | 61 | Finally, every program is threatened constantly by software patents. 62 | States should not allow patents to restrict development and use of 63 | software on general-purpose computers, but in those that do, we wish to 64 | avoid the special danger that patents applied to a free program could 65 | make it effectively proprietary. To prevent this, the GPL assures that 66 | patents cannot be used to render the program non-free. 67 | 68 | The precise terms and conditions for copying, distribution and 69 | modification follow. 70 | 71 | TERMS AND CONDITIONS 72 | 73 | 0. Definitions. 74 | 75 | "This License" refers to version 3 of the GNU General Public License. 76 | 77 | "Copyright" also means copyright-like laws that apply to other kinds of 78 | works, such as semiconductor masks. 79 | 80 | "The Program" refers to any copyrightable work licensed under this 81 | License. Each licensee is addressed as "you". "Licensees" and 82 | "recipients" may be individuals or organizations. 83 | 84 | To "modify" a work means to copy from or adapt all or part of the work 85 | in a fashion requiring copyright permission, other than the making of an 86 | exact copy. The resulting work is called a "modified version" of the 87 | earlier work or a work "based on" the earlier work. 88 | 89 | A "covered work" means either the unmodified Program or a work based 90 | on the Program. 91 | 92 | To "propagate" a work means to do anything with it that, without 93 | permission, would make you directly or secondarily liable for 94 | infringement under applicable copyright law, except executing it on a 95 | computer or modifying a private copy. Propagation includes copying, 96 | distribution (with or without modification), making available to the 97 | public, and in some countries other activities as well. 98 | 99 | To "convey" a work means any kind of propagation that enables other 100 | parties to make or receive copies. Mere interaction with a user through 101 | a computer network, with no transfer of a copy, is not conveying. 102 | 103 | An interactive user interface displays "Appropriate Legal Notices" 104 | to the extent that it includes a convenient and prominently visible 105 | feature that (1) displays an appropriate copyright notice, and (2) 106 | tells the user that there is no warranty for the work (except to the 107 | extent that warranties are provided), that licensees may convey the 108 | work under this License, and how to view a copy of this License. If 109 | the interface presents a list of user commands or options, such as a 110 | menu, a prominent item in the list meets this criterion. 111 | 112 | 1. Source Code. 113 | 114 | The "source code" for a work means the preferred form of the work 115 | for making modifications to it. "Object code" means any non-source 116 | form of a work. 117 | 118 | A "Standard Interface" means an interface that either is an official 119 | standard defined by a recognized standards body, or, in the case of 120 | interfaces specified for a particular programming language, one that 121 | is widely used among developers working in that language. 122 | 123 | The "System Libraries" of an executable work include anything, other 124 | than the work as a whole, that (a) is included in the normal form of 125 | packaging a Major Component, but which is not part of that Major 126 | Component, and (b) serves only to enable use of the work with that 127 | Major Component, or to implement a Standard Interface for which an 128 | implementation is available to the public in source code form. A 129 | "Major Component", in this context, means a major essential component 130 | (kernel, window system, and so on) of the specific operating system 131 | (if any) on which the executable work runs, or a compiler used to 132 | produce the work, or an object code interpreter used to run it. 133 | 134 | The "Corresponding Source" for a work in object code form means all 135 | the source code needed to generate, install, and (for an executable 136 | work) run the object code and to modify the work, including scripts to 137 | control those activities. However, it does not include the work's 138 | System Libraries, or general-purpose tools or generally available free 139 | programs which are used unmodified in performing those activities but 140 | which are not part of the work. For example, Corresponding Source 141 | includes interface definition files associated with source files for 142 | the work, and the source code for shared libraries and dynamically 143 | linked subprograms that the work is specifically designed to require, 144 | such as by intimate data communication or control flow between those 145 | subprograms and other parts of the work. 146 | 147 | The Corresponding Source need not include anything that users 148 | can regenerate automatically from other parts of the Corresponding 149 | Source. 150 | 151 | The Corresponding Source for a work in source code form is that 152 | same work. 153 | 154 | 2. Basic Permissions. 155 | 156 | All rights granted under this License are granted for the term of 157 | copyright on the Program, and are irrevocable provided the stated 158 | conditions are met. This License explicitly affirms your unlimited 159 | permission to run the unmodified Program. The output from running a 160 | covered work is covered by this License only if the output, given its 161 | content, constitutes a covered work. This License acknowledges your 162 | rights of fair use or other equivalent, as provided by copyright law. 163 | 164 | You may make, run and propagate covered works that you do not 165 | convey, without conditions so long as your license otherwise remains 166 | in force. You may convey covered works to others for the sole purpose 167 | of having them make modifications exclusively for you, or provide you 168 | with facilities for running those works, provided that you comply with 169 | the terms of this License in conveying all material for which you do 170 | not control copyright. Those thus making or running the covered works 171 | for you must do so exclusively on your behalf, under your direction 172 | and control, on terms that prohibit them from making any copies of 173 | your copyrighted material outside their relationship with you. 174 | 175 | Conveying under any other circumstances is permitted solely under 176 | the conditions stated below. Sublicensing is not allowed; section 10 177 | makes it unnecessary. 178 | 179 | 3. Protecting Users' Legal Rights From Anti-Circumvention Law. 180 | 181 | No covered work shall be deemed part of an effective technological 182 | measure under any applicable law fulfilling obligations under article 183 | 11 of the WIPO copyright treaty adopted on 20 December 1996, or 184 | similar laws prohibiting or restricting circumvention of such 185 | measures. 186 | 187 | When you convey a covered work, you waive any legal power to forbid 188 | circumvention of technological measures to the extent such circumvention 189 | is effected by exercising rights under this License with respect to 190 | the covered work, and you disclaim any intention to limit operation or 191 | modification of the work as a means of enforcing, against the work's 192 | users, your or third parties' legal rights to forbid circumvention of 193 | technological measures. 194 | 195 | 4. Conveying Verbatim Copies. 196 | 197 | You may convey verbatim copies of the Program's source code as you 198 | receive it, in any medium, provided that you conspicuously and 199 | appropriately publish on each copy an appropriate copyright notice; 200 | keep intact all notices stating that this License and any 201 | non-permissive terms added in accord with section 7 apply to the code; 202 | keep intact all notices of the absence of any warranty; and give all 203 | recipients a copy of this License along with the Program. 204 | 205 | You may charge any price or no price for each copy that you convey, 206 | and you may offer support or warranty protection for a fee. 207 | 208 | 5. Conveying Modified Source Versions. 209 | 210 | You may convey a work based on the Program, or the modifications to 211 | produce it from the Program, in the form of source code under the 212 | terms of section 4, provided that you also meet all of these conditions: 213 | 214 | a) The work must carry prominent notices stating that you modified 215 | it, and giving a relevant date. 216 | 217 | b) The work must carry prominent notices stating that it is 218 | released under this License and any conditions added under section 219 | 7. This requirement modifies the requirement in section 4 to 220 | "keep intact all notices". 221 | 222 | c) You must license the entire work, as a whole, under this 223 | License to anyone who comes into possession of a copy. This 224 | License will therefore apply, along with any applicable section 7 225 | additional terms, to the whole of the work, and all its parts, 226 | regardless of how they are packaged. This License gives no 227 | permission to license the work in any other way, but it does not 228 | invalidate such permission if you have separately received it. 229 | 230 | d) If the work has interactive user interfaces, each must display 231 | Appropriate Legal Notices; however, if the Program has interactive 232 | interfaces that do not display Appropriate Legal Notices, your 233 | work need not make them do so. 234 | 235 | A compilation of a covered work with other separate and independent 236 | works, which are not by their nature extensions of the covered work, 237 | and which are not combined with it such as to form a larger program, 238 | in or on a volume of a storage or distribution medium, is called an 239 | "aggregate" if the compilation and its resulting copyright are not 240 | used to limit the access or legal rights of the compilation's users 241 | beyond what the individual works permit. Inclusion of a covered work 242 | in an aggregate does not cause this License to apply to the other 243 | parts of the aggregate. 244 | 245 | 6. Conveying Non-Source Forms. 246 | 247 | You may convey a covered work in object code form under the terms 248 | of sections 4 and 5, provided that you also convey the 249 | machine-readable Corresponding Source under the terms of this License, 250 | in one of these ways: 251 | 252 | a) Convey the object code in, or embodied in, a physical product 253 | (including a physical distribution medium), accompanied by the 254 | Corresponding Source fixed on a durable physical medium 255 | customarily used for software interchange. 256 | 257 | b) Convey the object code in, or embodied in, a physical product 258 | (including a physical distribution medium), accompanied by a 259 | written offer, valid for at least three years and valid for as 260 | long as you offer spare parts or customer support for that product 261 | model, to give anyone who possesses the object code either (1) a 262 | copy of the Corresponding Source for all the software in the 263 | product that is covered by this License, on a durable physical 264 | medium customarily used for software interchange, for a price no 265 | more than your reasonable cost of physically performing this 266 | conveying of source, or (2) access to copy the 267 | Corresponding Source from a network server at no charge. 268 | 269 | c) Convey individual copies of the object code with a copy of the 270 | written offer to provide the Corresponding Source. This 271 | alternative is allowed only occasionally and noncommercially, and 272 | only if you received the object code with such an offer, in accord 273 | with subsection 6b. 274 | 275 | d) Convey the object code by offering access from a designated 276 | place (gratis or for a charge), and offer equivalent access to the 277 | Corresponding Source in the same way through the same place at no 278 | further charge. You need not require recipients to copy the 279 | Corresponding Source along with the object code. If the place to 280 | copy the object code is a network server, the Corresponding Source 281 | may be on a different server (operated by you or a third party) 282 | that supports equivalent copying facilities, provided you maintain 283 | clear directions next to the object code saying where to find the 284 | Corresponding Source. Regardless of what server hosts the 285 | Corresponding Source, you remain obligated to ensure that it is 286 | available for as long as needed to satisfy these requirements. 287 | 288 | e) Convey the object code using peer-to-peer transmission, provided 289 | you inform other peers where the object code and Corresponding 290 | Source of the work are being offered to the general public at no 291 | charge under subsection 6d. 292 | 293 | A separable portion of the object code, whose source code is excluded 294 | from the Corresponding Source as a System Library, need not be 295 | included in conveying the object code work. 296 | 297 | A "User Product" is either (1) a "consumer product", which means any 298 | tangible personal property which is normally used for personal, family, 299 | or household purposes, or (2) anything designed or sold for incorporation 300 | into a dwelling. In determining whether a product is a consumer product, 301 | doubtful cases shall be resolved in favor of coverage. For a particular 302 | product received by a particular user, "normally used" refers to a 303 | typical or common use of that class of product, regardless of the status 304 | of the particular user or of the way in which the particular user 305 | actually uses, or expects or is expected to use, the product. A product 306 | is a consumer product regardless of whether the product has substantial 307 | commercial, industrial or non-consumer uses, unless such uses represent 308 | the only significant mode of use of the product. 309 | 310 | "Installation Information" for a User Product means any methods, 311 | procedures, authorization keys, or other information required to install 312 | and execute modified versions of a covered work in that User Product from 313 | a modified version of its Corresponding Source. The information must 314 | suffice to ensure that the continued functioning of the modified object 315 | code is in no case prevented or interfered with solely because 316 | modification has been made. 317 | 318 | If you convey an object code work under this section in, or with, or 319 | specifically for use in, a User Product, and the conveying occurs as 320 | part of a transaction in which the right of possession and use of the 321 | User Product is transferred to the recipient in perpetuity or for a 322 | fixed term (regardless of how the transaction is characterized), the 323 | Corresponding Source conveyed under this section must be accompanied 324 | by the Installation Information. But this requirement does not apply 325 | if neither you nor any third party retains the ability to install 326 | modified object code on the User Product (for example, the work has 327 | been installed in ROM). 328 | 329 | The requirement to provide Installation Information does not include a 330 | requirement to continue to provide support service, warranty, or updates 331 | for a work that has been modified or installed by the recipient, or for 332 | the User Product in which it has been modified or installed. Access to a 333 | network may be denied when the modification itself materially and 334 | adversely affects the operation of the network or violates the rules and 335 | protocols for communication across the network. 336 | 337 | Corresponding Source conveyed, and Installation Information provided, 338 | in accord with this section must be in a format that is publicly 339 | documented (and with an implementation available to the public in 340 | source code form), and must require no special password or key for 341 | unpacking, reading or copying. 342 | 343 | 7. Additional Terms. 344 | 345 | "Additional permissions" are terms that supplement the terms of this 346 | License by making exceptions from one or more of its conditions. 347 | Additional permissions that are applicable to the entire Program shall 348 | be treated as though they were included in this License, to the extent 349 | that they are valid under applicable law. If additional permissions 350 | apply only to part of the Program, that part may be used separately 351 | under those permissions, but the entire Program remains governed by 352 | this License without regard to the additional permissions. 353 | 354 | When you convey a copy of a covered work, you may at your option 355 | remove any additional permissions from that copy, or from any part of 356 | it. (Additional permissions may be written to require their own 357 | removal in certain cases when you modify the work.) You may place 358 | additional permissions on material, added by you to a covered work, 359 | for which you have or can give appropriate copyright permission. 360 | 361 | Notwithstanding any other provision of this License, for material you 362 | add to a covered work, you may (if authorized by the copyright holders of 363 | that material) supplement the terms of this License with terms: 364 | 365 | a) Disclaiming warranty or limiting liability differently from the 366 | terms of sections 15 and 16 of this License; or 367 | 368 | b) Requiring preservation of specified reasonable legal notices or 369 | author attributions in that material or in the Appropriate Legal 370 | Notices displayed by works containing it; or 371 | 372 | c) Prohibiting misrepresentation of the origin of that material, or 373 | requiring that modified versions of such material be marked in 374 | reasonable ways as different from the original version; or 375 | 376 | d) Limiting the use for publicity purposes of names of licensors or 377 | authors of the material; or 378 | 379 | e) Declining to grant rights under trademark law for use of some 380 | trade names, trademarks, or service marks; or 381 | 382 | f) Requiring indemnification of licensors and authors of that 383 | material by anyone who conveys the material (or modified versions of 384 | it) with contractual assumptions of liability to the recipient, for 385 | any liability that these contractual assumptions directly impose on 386 | those licensors and authors. 387 | 388 | All other non-permissive additional terms are considered "further 389 | restrictions" within the meaning of section 10. If the Program as you 390 | received it, or any part of it, contains a notice stating that it is 391 | governed by this License along with a term that is a further 392 | restriction, you may remove that term. If a license document contains 393 | a further restriction but permits relicensing or conveying under this 394 | License, you may add to a covered work material governed by the terms 395 | of that license document, provided that the further restriction does 396 | not survive such relicensing or conveying. 397 | 398 | If you add terms to a covered work in accord with this section, you 399 | must place, in the relevant source files, a statement of the 400 | additional terms that apply to those files, or a notice indicating 401 | where to find the applicable terms. 402 | 403 | Additional terms, permissive or non-permissive, may be stated in the 404 | form of a separately written license, or stated as exceptions; 405 | the above requirements apply either way. 406 | 407 | 8. Termination. 408 | 409 | You may not propagate or modify a covered work except as expressly 410 | provided under this License. Any attempt otherwise to propagate or 411 | modify it is void, and will automatically terminate your rights under 412 | this License (including any patent licenses granted under the third 413 | paragraph of section 11). 414 | 415 | However, if you cease all violation of this License, then your 416 | license from a particular copyright holder is reinstated (a) 417 | provisionally, unless and until the copyright holder explicitly and 418 | finally terminates your license, and (b) permanently, if the copyright 419 | holder fails to notify you of the violation by some reasonable means 420 | prior to 60 days after the cessation. 421 | 422 | Moreover, your license from a particular copyright holder is 423 | reinstated permanently if the copyright holder notifies you of the 424 | violation by some reasonable means, this is the first time you have 425 | received notice of violation of this License (for any work) from that 426 | copyright holder, and you cure the violation prior to 30 days after 427 | your receipt of the notice. 428 | 429 | Termination of your rights under this section does not terminate the 430 | licenses of parties who have received copies or rights from you under 431 | this License. If your rights have been terminated and not permanently 432 | reinstated, you do not qualify to receive new licenses for the same 433 | material under section 10. 434 | 435 | 9. Acceptance Not Required for Having Copies. 436 | 437 | You are not required to accept this License in order to receive or 438 | run a copy of the Program. Ancillary propagation of a covered work 439 | occurring solely as a consequence of using peer-to-peer transmission 440 | to receive a copy likewise does not require acceptance. However, 441 | nothing other than this License grants you permission to propagate or 442 | modify any covered work. These actions infringe copyright if you do 443 | not accept this License. Therefore, by modifying or propagating a 444 | covered work, you indicate your acceptance of this License to do so. 445 | 446 | 10. Automatic Licensing of Downstream Recipients. 447 | 448 | Each time you convey a covered work, the recipient automatically 449 | receives a license from the original licensors, to run, modify and 450 | propagate that work, subject to this License. You are not responsible 451 | for enforcing compliance by third parties with this License. 452 | 453 | An "entity transaction" is a transaction transferring control of an 454 | organization, or substantially all assets of one, or subdividing an 455 | organization, or merging organizations. If propagation of a covered 456 | work results from an entity transaction, each party to that 457 | transaction who receives a copy of the work also receives whatever 458 | licenses to the work the party's predecessor in interest had or could 459 | give under the previous paragraph, plus a right to possession of the 460 | Corresponding Source of the work from the predecessor in interest, if 461 | the predecessor has it or can get it with reasonable efforts. 462 | 463 | You may not impose any further restrictions on the exercise of the 464 | rights granted or affirmed under this License. For example, you may 465 | not impose a license fee, royalty, or other charge for exercise of 466 | rights granted under this License, and you may not initiate litigation 467 | (including a cross-claim or counterclaim in a lawsuit) alleging that 468 | any patent claim is infringed by making, using, selling, offering for 469 | sale, or importing the Program or any portion of it. 470 | 471 | 11. Patents. 472 | 473 | A "contributor" is a copyright holder who authorizes use under this 474 | License of the Program or a work on which the Program is based. The 475 | work thus licensed is called the contributor's "contributor version". 476 | 477 | A contributor's "essential patent claims" are all patent claims 478 | owned or controlled by the contributor, whether already acquired or 479 | hereafter acquired, that would be infringed by some manner, permitted 480 | by this License, of making, using, or selling its contributor version, 481 | but do not include claims that would be infringed only as a 482 | consequence of further modification of the contributor version. For 483 | purposes of this definition, "control" includes the right to grant 484 | patent sublicenses in a manner consistent with the requirements of 485 | this License. 486 | 487 | Each contributor grants you a non-exclusive, worldwide, royalty-free 488 | patent license under the contributor's essential patent claims, to 489 | make, use, sell, offer for sale, import and otherwise run, modify and 490 | propagate the contents of its contributor version. 491 | 492 | In the following three paragraphs, a "patent license" is any express 493 | agreement or commitment, however denominated, not to enforce a patent 494 | (such as an express permission to practice a patent or covenant not to 495 | sue for patent infringement). To "grant" such a patent license to a 496 | party means to make such an agreement or commitment not to enforce a 497 | patent against the party. 498 | 499 | If you convey a covered work, knowingly relying on a patent license, 500 | and the Corresponding Source of the work is not available for anyone 501 | to copy, free of charge and under the terms of this License, through a 502 | publicly available network server or other readily accessible means, 503 | then you must either (1) cause the Corresponding Source to be so 504 | available, or (2) arrange to deprive yourself of the benefit of the 505 | patent license for this particular work, or (3) arrange, in a manner 506 | consistent with the requirements of this License, to extend the patent 507 | license to downstream recipients. "Knowingly relying" means you have 508 | actual knowledge that, but for the patent license, your conveying the 509 | covered work in a country, or your recipient's use of the covered work 510 | in a country, would infringe one or more identifiable patents in that 511 | country that you have reason to believe are valid. 512 | 513 | If, pursuant to or in connection with a single transaction or 514 | arrangement, you convey, or propagate by procuring conveyance of, a 515 | covered work, and grant a patent license to some of the parties 516 | receiving the covered work authorizing them to use, propagate, modify 517 | or convey a specific copy of the covered work, then the patent license 518 | you grant is automatically extended to all recipients of the covered 519 | work and works based on it. 520 | 521 | A patent license is "discriminatory" if it does not include within 522 | the scope of its coverage, prohibits the exercise of, or is 523 | conditioned on the non-exercise of one or more of the rights that are 524 | specifically granted under this License. You may not convey a covered 525 | work if you are a party to an arrangement with a third party that is 526 | in the business of distributing software, under which you make payment 527 | to the third party based on the extent of your activity of conveying 528 | the work, and under which the third party grants, to any of the 529 | parties who would receive the covered work from you, a discriminatory 530 | patent license (a) in connection with copies of the covered work 531 | conveyed by you (or copies made from those copies), or (b) primarily 532 | for and in connection with specific products or compilations that 533 | contain the covered work, unless you entered into that arrangement, 534 | or that patent license was granted, prior to 28 March 2007. 535 | 536 | Nothing in this License shall be construed as excluding or limiting 537 | any implied license or other defenses to infringement that may 538 | otherwise be available to you under applicable patent law. 539 | 540 | 12. No Surrender of Others' Freedom. 541 | 542 | If conditions are imposed on you (whether by court order, agreement or 543 | otherwise) that contradict the conditions of this License, they do not 544 | excuse you from the conditions of this License. If you cannot convey a 545 | covered work so as to satisfy simultaneously your obligations under this 546 | License and any other pertinent obligations, then as a consequence you may 547 | not convey it at all. For example, if you agree to terms that obligate you 548 | to collect a royalty for further conveying from those to whom you convey 549 | the Program, the only way you could satisfy both those terms and this 550 | License would be to refrain entirely from conveying the Program. 551 | 552 | 13. Use with the GNU Affero General Public License. 553 | 554 | Notwithstanding any other provision of this License, you have 555 | permission to link or combine any covered work with a work licensed 556 | under version 3 of the GNU Affero General Public License into a single 557 | combined work, and to convey the resulting work. The terms of this 558 | License will continue to apply to the part which is the covered work, 559 | but the special requirements of the GNU Affero General Public License, 560 | section 13, concerning interaction through a network will apply to the 561 | combination as such. 562 | 563 | 14. Revised Versions of this License. 564 | 565 | The Free Software Foundation may publish revised and/or new versions of 566 | the GNU General Public License from time to time. Such new versions will 567 | be similar in spirit to the present version, but may differ in detail to 568 | address new problems or concerns. 569 | 570 | Each version is given a distinguishing version number. If the 571 | Program specifies that a certain numbered version of the GNU General 572 | Public License "or any later version" applies to it, you have the 573 | option of following the terms and conditions either of that numbered 574 | version or of any later version published by the Free Software 575 | Foundation. If the Program does not specify a version number of the 576 | GNU General Public License, you may choose any version ever published 577 | by the Free Software Foundation. 578 | 579 | If the Program specifies that a proxy can decide which future 580 | versions of the GNU General Public License can be used, that proxy's 581 | public statement of acceptance of a version permanently authorizes you 582 | to choose that version for the Program. 583 | 584 | Later license versions may give you additional or different 585 | permissions. However, no additional obligations are imposed on any 586 | author or copyright holder as a result of your choosing to follow a 587 | later version. 588 | 589 | 15. Disclaimer of Warranty. 590 | 591 | THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY 592 | APPLICABLE LAW. EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT 593 | HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY 594 | OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, 595 | THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 596 | PURPOSE. THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM 597 | IS WITH YOU. SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF 598 | ALL NECESSARY SERVICING, REPAIR OR CORRECTION. 599 | 600 | 16. Limitation of Liability. 601 | 602 | IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING 603 | WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS 604 | THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY 605 | GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE 606 | USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF 607 | DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD 608 | PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS), 609 | EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF 610 | SUCH DAMAGES. 611 | 612 | 17. Interpretation of Sections 15 and 16. 613 | 614 | If the disclaimer of warranty and limitation of liability provided 615 | above cannot be given local legal effect according to their terms, 616 | reviewing courts shall apply local law that most closely approximates 617 | an absolute waiver of all civil liability in connection with the 618 | Program, unless a warranty or assumption of liability accompanies a 619 | copy of the Program in return for a fee. 620 | 621 | END OF TERMS AND CONDITIONS 622 | 623 | How to Apply These Terms to Your New Programs 624 | 625 | If you develop a new program, and you want it to be of the greatest 626 | possible use to the public, the best way to achieve this is to make it 627 | free software which everyone can redistribute and change under these terms. 628 | 629 | To do so, attach the following notices to the program. It is safest 630 | to attach them to the start of each source file to most effectively 631 | state the exclusion of warranty; and each file should have at least 632 | the "copyright" line and a pointer to where the full notice is found. 633 | 634 | 635 | Copyright (C) 636 | 637 | This program is free software: you can redistribute it and/or modify 638 | it under the terms of the GNU General Public License as published by 639 | the Free Software Foundation, either version 3 of the License, or 640 | (at your option) any later version. 641 | 642 | This program is distributed in the hope that it will be useful, 643 | but WITHOUT ANY WARRANTY; without even the implied warranty of 644 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 645 | GNU General Public License for more details. 646 | 647 | You should have received a copy of the GNU General Public License 648 | along with this program. If not, see . 649 | 650 | Also add information on how to contact you by electronic and paper mail. 651 | 652 | If the program does terminal interaction, make it output a short 653 | notice like this when it starts in an interactive mode: 654 | 655 | Copyright (C) 656 | This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'. 657 | This is free software, and you are welcome to redistribute it 658 | under certain conditions; type `show c' for details. 659 | 660 | The hypothetical commands `show w' and `show c' should show the appropriate 661 | parts of the General Public License. Of course, your program's commands 662 | might be different; for a GUI interface, you would use an "about box". 663 | 664 | You should also get your employer (if you work as a programmer) or school, 665 | if any, to sign a "copyright disclaimer" for the program, if necessary. 666 | For more information on this, and how to apply and follow the GNU GPL, see 667 | . 668 | 669 | The GNU General Public License does not permit incorporating your program 670 | into proprietary programs. If your program is a subroutine library, you 671 | may consider it more useful to permit linking proprietary applications with 672 | the library. If this is what you want to do, use the GNU Lesser General 673 | Public License instead of this License. But first, please read 674 | . 675 | -------------------------------------------------------------------------------- /stockfish/13/INSTALL_RECEIPT.json: -------------------------------------------------------------------------------- 1 | {"homebrew_version":"3.0.2-23-g9b94234","used_options":[],"unused_options":[],"built_as_bottle":true,"poured_from_bottle":true,"installed_as_dependency":false,"installed_on_request":true,"changed_files":["INSTALL_RECEIPT.json"],"time":1615694529,"source_modified_time":1613682895,"HEAD":null,"stdlib":null,"compiler":"clang","aliases":[],"runtime_dependencies":[],"source":{"path":"/usr/local/Homebrew/Library/Taps/homebrew/homebrew-core/Formula/stockfish.rb","tap":"homebrew/core","spec":"stable","versions":{"stable":"13","head":"HEAD","version_scheme":0}},"arch":"x86_64","built_on":{"os":"Macintosh","os_version":"macOS 10.15.7","cpu_family":"penryn","xcode":"12.4","clt":"12.4.0.0.1.1610135815"}} -------------------------------------------------------------------------------- /stockfish/13/README.md: -------------------------------------------------------------------------------- 1 | ## Overview 2 | 3 | [![Build Status](https://travis-ci.org/official-stockfish/Stockfish.svg?branch=master)](https://travis-ci.org/official-stockfish/Stockfish) 4 | [![Build Status](https://ci.appveyor.com/api/projects/status/github/official-stockfish/Stockfish?branch=master&svg=true)](https://ci.appveyor.com/project/mcostalba/stockfish/branch/master) 5 | 6 | [Stockfish](https://stockfishchess.org) is a free, powerful UCI chess engine 7 | derived from Glaurung 2.1. Stockfish is not a complete chess program and requires a 8 | UCI-compatible graphical user interface (GUI) (e.g. XBoard with PolyGlot, Scid, 9 | Cute Chess, eboard, Arena, Sigma Chess, Shredder, Chess Partner or Fritz) in order 10 | to be used comfortably. Read the documentation for your GUI of choice for information 11 | about how to use Stockfish with it. 12 | 13 | The Stockfish engine features two evaluation functions for chess, the classical 14 | evaluation based on handcrafted terms, and the NNUE evaluation based on efficiently 15 | updatable neural networks. The classical evaluation runs efficiently on almost all 16 | CPU architectures, while the NNUE evaluation benefits from the vector 17 | intrinsics available on most CPUs (sse2, avx2, neon, or similar). 18 | 19 | 20 | ## Files 21 | 22 | This distribution of Stockfish consists of the following files: 23 | 24 | * Readme.md, the file you are currently reading. 25 | 26 | * Copying.txt, a text file containing the GNU General Public License version 3. 27 | 28 | * AUTHORS, a text file with the list of authors for the project 29 | 30 | * src, a subdirectory containing the full source code, including a Makefile 31 | that can be used to compile Stockfish on Unix-like systems. 32 | 33 | * a file with the .nnue extension, storing the neural network for the NNUE 34 | evaluation. Binary distributions will have this file embedded. 35 | 36 | ## UCI options 37 | 38 | Currently, Stockfish has the following UCI options: 39 | 40 | * #### Threads 41 | The number of CPU threads used for searching a position. For best performance, set 42 | this equal to the number of CPU cores available. 43 | 44 | * #### Hash 45 | The size of the hash table in MB. It is recommended to set Hash after setting Threads. 46 | 47 | * #### Clear Hash 48 | Clear the hash table. 49 | 50 | * #### Ponder 51 | Let Stockfish ponder its next move while the opponent is thinking. 52 | 53 | * #### MultiPV 54 | Output the N best lines (principal variations, PVs) when searching. 55 | Leave at 1 for best performance. 56 | 57 | * #### Use NNUE 58 | Toggle between the NNUE and classical evaluation functions. If set to "true", 59 | the network parameters must be available to load from file (see also EvalFile), 60 | if they are not embedded in the binary. 61 | 62 | * #### EvalFile 63 | The name of the file of the NNUE evaluation parameters. Depending on the GUI the 64 | filename might have to include the full path to the folder/directory that contains the file. 65 | Other locations, such as the directory that contains the binary and the working directory, 66 | are also searched. 67 | 68 | * #### UCI_AnalyseMode 69 | An option handled by your GUI. 70 | 71 | * #### UCI_Chess960 72 | An option handled by your GUI. If true, Stockfish will play Chess960. 73 | 74 | * #### UCI_ShowWDL 75 | If enabled, show approximate WDL statistics as part of the engine output. 76 | These WDL numbers model expected game outcomes for a given evaluation and 77 | game ply for engine self-play at fishtest LTC conditions (60+0.6s per game). 78 | 79 | * #### UCI_LimitStrength 80 | Enable weaker play aiming for an Elo rating as set by UCI_Elo. This option overrides Skill Level. 81 | 82 | * #### UCI_Elo 83 | If enabled by UCI_LimitStrength, aim for an engine strength of the given Elo. 84 | This Elo rating has been calibrated at a time control of 60s+0.6s and anchored to CCRL 40/4. 85 | 86 | * #### Skill Level 87 | Lower the Skill Level in order to make Stockfish play weaker (see also UCI_LimitStrength). 88 | Internally, MultiPV is enabled, and with a certain probability depending on the Skill Level a 89 | weaker move will be played. 90 | 91 | * #### SyzygyPath 92 | Path to the folders/directories storing the Syzygy tablebase files. Multiple 93 | directories are to be separated by ";" on Windows and by ":" on Unix-based 94 | operating systems. Do not use spaces around the ";" or ":". 95 | 96 | Example: `C:\tablebases\wdl345;C:\tablebases\wdl6;D:\tablebases\dtz345;D:\tablebases\dtz6` 97 | 98 | It is recommended to store .rtbw files on an SSD. There is no loss in storing 99 | the .rtbz files on a regular HD. It is recommended to verify all md5 checksums 100 | of the downloaded tablebase files (`md5sum -c checksum.md5`) as corruption will 101 | lead to engine crashes. 102 | 103 | * #### SyzygyProbeDepth 104 | Minimum remaining search depth for which a position is probed. Set this option 105 | to a higher value to probe less aggressively if you experience too much slowdown 106 | (in terms of nps) due to tablebase probing. 107 | 108 | * #### Syzygy50MoveRule 109 | Disable to let fifty-move rule draws detected by Syzygy tablebase probes count 110 | as wins or losses. This is useful for ICCF correspondence games. 111 | 112 | * #### SyzygyProbeLimit 113 | Limit Syzygy tablebase probing to positions with at most this many pieces left 114 | (including kings and pawns). 115 | 116 | * #### Contempt 117 | A positive value for contempt favors middle game positions and avoids draws, 118 | effective for the classical evaluation only. 119 | 120 | * #### Analysis Contempt 121 | By default, contempt is set to prefer the side to move. Set this option to "White" 122 | or "Black" to analyse with contempt for that side, or "Off" to disable contempt. 123 | 124 | * #### Move Overhead 125 | Assume a time delay of x ms due to network and GUI overheads. This is useful to 126 | avoid losses on time in those cases. 127 | 128 | * #### Slow Mover 129 | Lower values will make Stockfish take less time in games, higher values will 130 | make it think longer. 131 | 132 | * #### nodestime 133 | Tells the engine to use nodes searched instead of wall time to account for 134 | elapsed time. Useful for engine testing. 135 | 136 | * #### Debug Log File 137 | Write all communication to and from the engine into a text file. 138 | 139 | ## A note on classical evaluation versus NNUE evaluation 140 | 141 | Both approaches assign a value to a position that is used in alpha-beta (PVS) search 142 | to find the best move. The classical evaluation computes this value as a function 143 | of various chess concepts, handcrafted by experts, tested and tuned using fishtest. 144 | The NNUE evaluation computes this value with a neural network based on basic 145 | inputs (e.g. piece positions only). The network is optimized and trained 146 | on the evaluations of millions of positions at moderate search depth. 147 | 148 | The NNUE evaluation was first introduced in shogi, and ported to Stockfish afterward. 149 | It can be evaluated efficiently on CPUs, and exploits the fact that only parts 150 | of the neural network need to be updated after a typical chess move. 151 | [The nodchip repository](https://github.com/nodchip/Stockfish) provides additional 152 | tools to train and develop the NNUE networks. On CPUs supporting modern vector instructions 153 | (avx2 and similar), the NNUE evaluation results in much stronger playing strength, even 154 | if the nodes per second computed by the engine is somewhat lower (roughly 80% of nps 155 | is typical). 156 | 157 | Notes: 158 | 159 | 1) the NNUE evaluation depends on the Stockfish binary and the network parameter 160 | file (see the EvalFile UCI option). Not every parameter file is compatible with a given 161 | Stockfish binary, but the default value of the EvalFile UCI option is the name of a network 162 | that is guaranteed to be compatible with that binary. 163 | 164 | 2) to use the NNUE evaluation, the additional data file with neural network parameters 165 | needs to be available. Normally, this file is already embedded in the binary or it 166 | can be downloaded. The filename for the default (recommended) net can be found as the default 167 | value of the `EvalFile` UCI option, with the format `nn-[SHA256 first 12 digits].nnue` 168 | (for instance, `nn-c157e0a5755b.nnue`). This file can be downloaded from 169 | ``` 170 | https://tests.stockfishchess.org/api/nn/[filename] 171 | ``` 172 | replacing `[filename]` as needed. 173 | 174 | ## What to expect from the Syzygy tablebases? 175 | 176 | If the engine is searching a position that is not in the tablebases (e.g. 177 | a position with 8 pieces), it will access the tablebases during the search. 178 | If the engine reports a very large score (typically 153.xx), this means 179 | it has found a winning line into a tablebase position. 180 | 181 | If the engine is given a position to search that is in the tablebases, it 182 | will use the tablebases at the beginning of the search to preselect all 183 | good moves, i.e. all moves that preserve the win or preserve the draw while 184 | taking into account the 50-move rule. 185 | It will then perform a search only on those moves. **The engine will not move 186 | immediately**, unless there is only a single good move. **The engine likely 187 | will not report a mate score, even if the position is known to be won.** 188 | 189 | It is therefore clear that this behaviour is not identical to what one might 190 | be used to with Nalimov tablebases. There are technical reasons for this 191 | difference, the main technical reason being that Nalimov tablebases use the 192 | DTM metric (distance-to-mate), while the Syzygy tablebases use a variation of the 193 | DTZ metric (distance-to-zero, zero meaning any move that resets the 50-move 194 | counter). This special metric is one of the reasons that the Syzygy tablebases are 195 | more compact than Nalimov tablebases, while still storing all information 196 | needed for optimal play and in addition being able to take into account 197 | the 50-move rule. 198 | 199 | ## Large Pages 200 | 201 | Stockfish supports large pages on Linux and Windows. Large pages make 202 | the hash access more efficient, improving the engine speed, especially 203 | on large hash sizes. Typical increases are 5..10% in terms of nodes per 204 | second, but speed increases up to 30% have been measured. The support is 205 | automatic. Stockfish attempts to use large pages when available and 206 | will fall back to regular memory allocation when this is not the case. 207 | 208 | ### Support on Linux 209 | 210 | Large page support on Linux is obtained by the Linux kernel 211 | transparent huge pages functionality. Typically, transparent huge pages 212 | are already enabled, and no configuration is needed. 213 | 214 | ### Support on Windows 215 | 216 | The use of large pages requires "Lock Pages in Memory" privilege. See 217 | [Enable the Lock Pages in Memory Option (Windows)](https://docs.microsoft.com/en-us/sql/database-engine/configure-windows/enable-the-lock-pages-in-memory-option-windows) 218 | on how to enable this privilege, then run [RAMMap](https://docs.microsoft.com/en-us/sysinternals/downloads/rammap) 219 | to double-check that large pages are used. We suggest that you reboot 220 | your computer after you have enabled large pages, because long Windows 221 | sessions suffer from memory fragmentation, which may prevent Stockfish 222 | from getting large pages: a fresh session is better in this regard. 223 | 224 | ## Compiling Stockfish yourself from the sources 225 | 226 | Stockfish has support for 32 or 64-bit CPUs, certain hardware 227 | instructions, big-endian machines such as Power PC, and other platforms. 228 | 229 | On Unix-like systems, it should be easy to compile Stockfish 230 | directly from the source code with the included Makefile in the folder 231 | `src`. In general it is recommended to run `make help` to see a list of make 232 | targets with corresponding descriptions. 233 | 234 | ``` 235 | cd src 236 | make help 237 | make net 238 | make build ARCH=x86-64-modern 239 | ``` 240 | 241 | When not using the Makefile to compile (for instance, with Microsoft MSVC) you 242 | need to manually set/unset some switches in the compiler command line; see 243 | file *types.h* for a quick reference. 244 | 245 | When reporting an issue or a bug, please tell us which version and 246 | compiler you used to create your executable. These informations can 247 | be found by typing the following commands in a console: 248 | 249 | ``` 250 | ./stockfish compiler 251 | ``` 252 | 253 | ## Understanding the code base and participating in the project 254 | 255 | Stockfish's improvement over the last couple of years has been a great 256 | community effort. There are a few ways to help contribute to its growth. 257 | 258 | ### Donating hardware 259 | 260 | Improving Stockfish requires a massive amount of testing. You can donate 261 | your hardware resources by installing the [Fishtest Worker](https://github.com/glinscott/fishtest/wiki/Running-the-worker:-overview) 262 | and view the current tests on [Fishtest](https://tests.stockfishchess.org/tests). 263 | 264 | ### Improving the code 265 | 266 | If you want to help improve the code, there are several valuable resources: 267 | 268 | * [In this wiki,](https://www.chessprogramming.org) many techniques used in 269 | Stockfish are explained with a lot of background information. 270 | 271 | * [The section on Stockfish](https://www.chessprogramming.org/Stockfish) 272 | describes many features and techniques used by Stockfish. However, it is 273 | generic rather than being focused on Stockfish's precise implementation. 274 | Nevertheless, a helpful resource. 275 | 276 | * The latest source can always be found on [GitHub](https://github.com/official-stockfish/Stockfish). 277 | Discussions about Stockfish take place these days mainly in the [FishCooking](https://groups.google.com/forum/#!forum/fishcooking) 278 | group and on the [Stockfish Discord channel](https://discord.gg/nv8gDtt). 279 | The engine testing is done on [Fishtest](https://tests.stockfishchess.org/tests). 280 | If you want to help improve Stockfish, please read this [guideline](https://github.com/glinscott/fishtest/wiki/Creating-my-first-test) 281 | first, where the basics of Stockfish development are explained. 282 | 283 | 284 | ## Terms of use 285 | 286 | Stockfish is free, and distributed under the **GNU General Public License version 3** 287 | (GPL v3). Essentially, this means you are free to do almost exactly 288 | what you want with the program, including distributing it among your 289 | friends, making it available for download from your website, selling 290 | it (either by itself or as part of some bigger software package), or 291 | using it as the starting point for a software project of your own. 292 | 293 | The only real limitation is that whenever you distribute Stockfish in 294 | some way, you MUST always include the full source code, or a pointer 295 | to where the source code can be found, to generate the exact binary 296 | you are distributing. If you make any changes to the source code, 297 | these changes must also be made available under the GPL. 298 | 299 | For full details, read the copy of the GPL v3 found in the file named 300 | *Copying.txt*. 301 | -------------------------------------------------------------------------------- /stockfish/13/bin/stockfish: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/realnihal/Chess-AI-with-TensorFlow/2ecda18b164ec63fb38cf97b919845e595b83df0/stockfish/13/bin/stockfish --------------------------------------------------------------------------------