File size: 18,214 Bytes
4ca4e4c | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 | import numpy as np
import torch
import string
import torch.nn.functional as F
import random
import math
from collections import defaultdict
from transformers import AutoTokenizer
#################################################################
# Tokenizer
class Tokenizer:
def __init__(self, TO_TOKEN, vocab_tokens, number_tokens):
self.TO_TOKEN = TO_TOKEN
self.TO_STR = {v:k for k, v in TO_TOKEN.items()}
self.vocab = np.array(list(TO_TOKEN.keys()))
self.vocab_tokens = vocab_tokens
self.number_tokens = number_tokens
self.num_vocab = len(self.vocab_tokens)
self.num_numbers = len(self.number_tokens)
self.bos_token = self.TO_TOKEN['<bos>']
self.eos_token = self.TO_TOKEN['<eos>']
self.null = '<null>'
# Human readible printing
vocab_part = {self.TO_TOKEN[k]: v for (k, v) in zip(self.vocab_tokens, string.ascii_lowercase[:self.num_vocab])}
number_part = {self.TO_TOKEN[t]: t[1:] for t in self.number_tokens}
self.TO_STRING = {**vocab_part, **number_part}
self.TO_STRING[self.bos_token] = "$"
self.TO_STRING[self.eos_token] = "."
self.TO_STRING[self.TO_TOKEN[self.null]] = "_"
def __call__(self, x):
encoded = [self.TO_TOKEN[c] for c in x]
return torch.tensor(encoded, dtype=torch.int64)
def decode(self, x):
x = x.detach().cpu().numpy()
decoded = [str(t) if t not in self.TO_STR else self.TO_STR[t] for t in x]
return decoded
def __len__(self):
return len(self.TO_TOKEN)
def to_string(self, x, pytorch=True):
if pytorch:
return "".join([self.TO_STRING[t.item()] for t in x])
else:
return "".join([self.TO_STRING[self.TO_TOKEN[t]] for t in x])
def get_tokenizer(args):
if args.model == "pretrained":
tokenizer = AutoTokenizer.from_pretrained(args.pretrained_model)
return tokenizer
vocab_tokens = ["V%d" % i for i in range(args.num_vocab)]
if args.train_task in ["var-copy", "var-copy-rep"]:
number_tokens = ["#%d" % (5+i) for i in range(args.num_numbers)]
else:
number_tokens = ["#%d" % i for i in range(args.num_numbers)]
vocab = vocab_tokens + number_tokens + ["<bos>", "<eos>", "<null>"]
TO_TOKEN = dict(zip(vocab, range(len(vocab))))
tokenizer = Tokenizer(TO_TOKEN, vocab_tokens, number_tokens)
return tokenizer
#################################################################
# Sequence Generation
def rand_seq(tokenizer, length, num_vocab, num_numbers, p_numbers=-1):
if p_numbers == -1:
p_numbers = num_numbers / (num_vocab + num_numbers)
if num_numbers != 0:
props = {"V": (1-p_numbers)/num_vocab, "#": p_numbers/num_numbers, "<": 0}
else:
props = {"V": 1/num_vocab, "#": 0, "<": 0}
props = np.array([props[i[0]] for i in tokenizer.vocab])
return np.random.choice(tokenizer.vocab, size=length, p=props).tolist()
# For other special generations
def rand_seq_special(tokenizer, length, num_vocab, num_numbers, p_numbers=-1, special_type=None):
if special_type == "repetitive_vocab":
if p_numbers == -1:
p_numbers = num_numbers / (num_vocab + num_numbers)
if num_numbers != 0:
props = {"V": 0, "#": p_numbers/num_numbers, "<": 0}
else:
props = {"V": 0, "#": 0, "<": 0}
if num_numbers != 0:
props_V0 = (1-p_numbers)
else:
props_V0 = 1
props = np.array([props[i[0]] if i != "V0" else props_V0 for i in tokenizer.vocab])
tile_length = 3
props_tile = {"V": 1./num_vocab, "#": 0, "<": 0}
props_tile = np.array([props_tile[i[0]] for i in tokenizer.vocab])
ret_seq = np.random.choice(tokenizer.vocab, size=length, p=props)
ret_seq2 = np.tile(np.random.choice(tokenizer.vocab, size=tile_length, p=props_tile), (length // tile_length + 1))[:length]
return np.where(ret_seq == "V0", ret_seq2, ret_seq).tolist()
else:
assert False, "Not implemented"
def force_args(args):
if args.train_task == "var-copy":
pass
if args.train_task == "var-copy-rep":
pass
if args.train_task in ["decode-recall", "decode-recall-last"]:
args.num_numbers = 2
args.num_vocab = int(2 ** math.floor(math.log(args.num_vocab) / math.log(2)))
if args.train_task == "assoc-recall":
args.num_numbers = 0
if args.train_task == "assoc-recall-mk":
size_key = 2
args.num_numbers = 0
args.num_vocab = 1 + int(args.num_vocab ** (1./size_key))
if args.train_task == "addition":
args.num_numbers = 10 # Decimal addition
args.num_vocab = 2 # For +, =
args.min_train_length = 3*args.min_train_length+3
args.max_train_length = 3*args.max_train_length+3
args.min_eval_length = 3*args.min_eval_length+3
args.max_eval_length = 3*args.max_eval_length+3
task_choices = ["var-copy", "var-copy-rep", "decode-recall", "decode-recall-last", "assoc-recall", "assoc-recall-mk", "addition"]
def generate_seq_and_mask(tokenizer, length, task, p=0.2):
num_vocab = tokenizer.num_vocab
num_numbers = tokenizer.num_numbers
if task == "var-copy":
# Start with num_numbers vocab tokens
input_seq = rand_seq(tokenizer, length, num_vocab, num_numbers, p_numbers=p)
# input_seq = rand_seq(tokenizer, length, num_vocab, num_numbers)
nums = [(i, int(c[1:])) for (i, c) in enumerate(input_seq) if c in tokenizer.number_tokens]
# The real task, if not degenerate
if len(nums) > 0:
output_seq = ["<null>"] * nums[0][0]
for i in range(len(nums)-1):
if nums[i][0]-nums[i][1] < 0:
if nums[i+1][0]-nums[i][1] < 0:
output_seq += ["<null>"] * (nums[i+1][0]-nums[i][0])
else:
output_seq += ["<null>"] * (nums[i][1]-nums[i][0])
output_seq += input_seq[:nums[i+1][0]-nums[i][1]]
else:
output_seq += input_seq[nums[i][0]-nums[i][1]:nums[i+1][0]-nums[i][1]]
if nums[-1][0]-nums[-1][1] < 0:
output_seq += ["<null>"] * (nums[-1][1]-nums[-1][0])
output_seq += input_seq[:-nums[-1][1]]
else:
output_seq += input_seq[nums[-1][0]-nums[-1][1]:-nums[-1][1]]
else:
output_seq = ["<null>"] * length
input_seq = ["<bos>"] + input_seq + ["<eos>"]
output_seq = ["<bos>"] + output_seq + ["<eos>"]
# output_seq = ["<bos>"] + output_seq[:length] + ["<eos>"]
elif task == "var-copy-rep":
# Start with num_numbers vocab tokens
# input_seq = rand_seq(tokenizer, length, num_vocab, num_numbers, p_numbers=p)
input_seq = rand_seq_special(tokenizer, length, num_vocab, num_numbers, p_numbers=p, special_type="repetitive_vocab")
nums = [(i, int(c[1:])) for (i, c) in enumerate(input_seq) if c in tokenizer.number_tokens]
# The real task, if not degenerate
if len(nums) > 0:
output_seq = ["<null>"] * nums[0][0]
for i in range(len(nums)-1):
if nums[i][0]-nums[i][1] < 0:
if nums[i+1][0]-nums[i][1] < 0:
output_seq += ["<null>"] * (nums[i+1][0]-nums[i][0])
else:
output_seq += ["<null>"] * (nums[i][1]-nums[i][0])
output_seq += input_seq[:nums[i+1][0]-nums[i][1]]
else:
output_seq += input_seq[nums[i][0]-nums[i][1]:nums[i+1][0]-nums[i][1]]
if nums[-1][0]-nums[-1][1] < 0:
output_seq += ["<null>"] * (nums[-1][1]-nums[-1][0])
output_seq += input_seq[:-nums[-1][1]]
else:
output_seq += input_seq[nums[-1][0]-nums[-1][1]:-nums[-1][1]]
else:
output_seq = ["<null>"] * length
input_seq = ["<bos>"] + input_seq + ["<eos>"]
output_seq = ["<bos>"] + output_seq + ["<eos>"]
# output_seq = ["<bos>"] + output_seq[:length] + ["<eos>"]
elif task == "decode-recall":
input_seq = rand_seq(tokenizer, length, num_vocab, num_numbers, p_numbers=p)
output_seq = [None for _ in range(len(input_seq))]
assoc = {v: "<null>" for v in tokenizer.vocab}
s = 0
for i in range(len(output_seq)):
if i != 0:
assoc[input_seq[i-1]] = input_seq[i]
if input_seq[i][0] == '#':
# s = (2 * s + int(input_seq[i][1:])) % num_numbers
s = (2 * s + int(input_seq[i][1:])) % num_vocab
# if i-s < 0:
# output_seq[i] = "<null>"
# else:
# output_seq[i] = input_seq[i-s]
output_seq[i] = assoc["V%d" % s]
elif task == "decode-recall-last":
input_seq = rand_seq(tokenizer, length, num_vocab, num_numbers, p_numbers=0)
output_seq = ["<null>" for _ in range(len(input_seq))]
n_bits = int(math.log(num_vocab)/math.log(2))
target = np.random.randint(0, num_vocab)
temp = target
for i in range(length-1, length-1-n_bits, -1):
input_seq[i] = "#%d" % (temp % 2)
temp = temp // 2
try:
i = length-2-n_bits - input_seq[-2-n_bits::-1].index("V%d" % target)
output_seq[-1] = input_seq[i+1]
except ValueError:
pass
elif task == "assoc-recall":
input_seq = rand_seq(tokenizer, length, num_vocab, num_numbers, p_numbers=0.2)
output_seq = [None for _ in range(len(input_seq))]
assoc = {v: "<null>" for v in tokenizer.vocab}
for i in range(len(output_seq)):
if i != 0:
assoc[input_seq[i-1]] = input_seq[i]
output_seq[i] = assoc[input_seq[i]]
elif task == "assoc-recall-mk":
size_key = 2
input_seq = rand_seq(tokenizer, length, num_vocab, 0, p_numbers=0.2)
output_seq = ["<null>" for _ in range(len(input_seq))]
assoc = defaultdict(lambda: "<null>")
for i in range(len(output_seq)):
if i > size_key:
key = tuple(input_seq[i-size_key:i])
assoc[key] = input_seq[i]
if i+1 > size_key:
key = tuple(input_seq[i-size_key+1:i+1])
output_seq[i] = assoc[key]
elif task == "addition":
len_number = (length+1) // 3 - 1
max_num = num_numbers ** len_number
num1 = np.random.randint(0, max_num)
num2 = np.random.randint(0, max_num)
num3 = (num1 + num2) % max_num
input_seq = ["<null>" for _ in range(length)]
output_seq = ["<null>" for _ in range(length)]
for i in range(len_number-1, -1, -1):
input_seq[i] = "#%d" % (num1 % num_numbers)
num1 = num1 // num_numbers
input_seq[len_number] = "V0"
for i in range(2*len_number, len_number, -1):
input_seq[i] = "#%d" % (num2 % num_numbers)
num2 = num2 // num_numbers
input_seq[2*len_number+1] = "V1"
for i in range(3*len_number+1, 2*len_number+1, -1):
input_seq[i] = "#%d" % (num3 % num_numbers)
output_seq[i-1] = "#%d" % (num3 % num_numbers)
num3 = num3 // num_numbers
else:
print("Task name:", task)
assert False # Not implemented
# Create the mask
mask = [0 if i in ["<bos>", "<eos>", "<null>"] else 1 for i in output_seq]
return input_seq, output_seq, mask
#################################################################
# Datasets
class Dataset:
def __init__(self,
tokenizer,
train_task="var_copy",
sequence_length=220,
min_subseq_length=20,
max_subseq_length=50,
num_examples=1000,
batch_size=8,
p=0.2):
self.tokenizer = tokenizer
self.train_task = train_task
self.num_vocab = self.tokenizer.num_vocab
self.num_numbers = self.tokenizer.num_numbers
self.sequence_length = sequence_length
self.min_subseq_length = min_subseq_length
self.max_subseq_length = max_subseq_length
self.num_examples = num_examples
self.batch_size = batch_size
self.p = p
def __len__(self):
return self.num_examples
def __getitem__(self, idx):
batch = {'input': [], 'input_ids': [], 'output': [], 'output_ids': [], 'mask': []}
for _ in range(self.batch_size):
# Fill the context with subsequences of the desired task
prospective_len = 0
input_seq = []
output_seq = []
mask = []
while prospective_len < self.sequence_length:
# Sample for the task
length = np.random.randint(self.min_subseq_length, self.max_subseq_length+1)
input_sample, output_sample, mask_sample = generate_seq_and_mask(self.tokenizer, length, self.train_task, self.p)
# Add the sample to the context
if prospective_len + len(input_sample) <= self.sequence_length:
prospective_len += len(input_sample)
input_seq += input_sample
output_seq += output_sample
mask += mask_sample
# Not enough room for another sample
else:
remaining_len = self.sequence_length - prospective_len
remaining_mask_len = self.sequence_length - prospective_len
input_seq += input_sample[:remaining_len]
output_seq += output_sample[:remaining_len]
mask += [0] * (remaining_mask_len) # Just mask it
break
# Add the sequence to the sampled dataset
assert len(input_seq) == len(mask)
input_ids = self.tokenizer(input_seq)
output_ids = self.tokenizer(output_seq)
mask = torch.tensor(mask)
batch['input'].append(input_seq)
batch['input_ids'].append(input_ids)
batch['output'].append(output_seq)
batch['output_ids'].append(output_ids)
batch['mask'].append(mask)
batch['input_ids'] = torch.stack(batch['input_ids'], dim=0)
batch['output_ids'] = torch.stack(batch['output_ids'], dim=0)
batch['mask'] = torch.stack(batch['mask'], dim=0)
return batch
class EvalDataset:
def __init__(self,
tokenizer,
train_task="var_copy",
sequence_length=220,
min_subseq_length=20,
max_subseq_length=50,
num_examples=1000,
batch_size=8,
p=0.2):
self.tokenizer = tokenizer
self.train_task = train_task
self.sequence_length = sequence_length
self.min_subseq_length = min_subseq_length
self.max_subseq_length = max_subseq_length
self.num_examples = num_examples
self.batch_size = batch_size
self.p = p
def __len__(self):
return self.num_examples
def __getitem__(self, idx):
batch = {'input': [], 'input_ids': [], 'output': [], 'output_ids': [], 'mask': []}
for _ in range(self.batch_size):
# Fill the context with subsequences of the desired task
prospective_len = 0
input_seq = []
output_seq = []
mask = []
# Sample for the task
length = np.random.randint(self.min_subseq_length, self.max_subseq_length+1)
input_seq, output_seq, mask = generate_seq_and_mask(self.tokenizer, length, self.train_task, self.p)
# DO NOT REPLACE
# Fill the context with null tokens
input_seq += ["<null>"] * (self.sequence_length - len(input_seq))
output_seq += ["<null>"] * (self.sequence_length - len(output_seq))
mask += [0] * (self.sequence_length - len(mask))
# Add the sequence to the sampled dataset
assert len(input_seq) == len(mask)
input_ids = self.tokenizer(input_seq)
output_ids = self.tokenizer(output_seq)
mask = torch.tensor(mask)
batch['input'].append(input_seq)
batch['input_ids'].append(input_ids)
batch['output'].append(output_seq)
batch['output_ids'].append(output_ids)
batch['mask'].append(mask)
batch['input_ids'] = torch.stack(batch['input_ids'], dim=0)
batch['output_ids'] = torch.stack(batch['output_ids'], dim=0)
batch['mask'] = torch.stack(batch['mask'], dim=0)
return batch
#################################################################
# Util functions
def get_train_dataset(args, tokenizer):
train_dataset = Dataset(
tokenizer=tokenizer,
train_task=args.train_task,
sequence_length=args.sequence_length,
min_subseq_length=args.min_train_length,
max_subseq_length=args.max_train_length,
num_examples=args.num_examples,
batch_size=args.train_batch_size,
p=args.p
)
return train_dataset
def get_eval_dataset(args, tokenizer, min_length, max_length):
eval_dataset = EvalDataset(
tokenizer=tokenizer,
train_task=args.train_task,
sequence_length=args.sequence_length,
min_subseq_length=min_length,
max_subseq_length=max_length,
num_examples=args.num_examples,
batch_size=args.eval_batch_size,
p=args.p
)
return eval_dataset
|