add streaming dataset support for pretraining datasets
Browse files- README.md +2 -0
- scripts/finetune.py +3 -20
- src/axolotl/utils/data.py +112 -28
- src/axolotl/utils/validation.py +5 -0
- tests/test_validation.py +51 -0
README.md
CHANGED
@@ -410,6 +410,8 @@ optimizer:
|
|
410 |
# specify weight decay
|
411 |
weight_decay:
|
412 |
|
|
|
|
|
413 |
# whether to use xformers attention patch https://github.com/facebookresearch/xformers:
|
414 |
xformers_attention:
|
415 |
# whether to use flash attention patch https://github.com/HazyResearch/flash-attention:
|
|
|
410 |
# specify weight decay
|
411 |
weight_decay:
|
412 |
|
413 |
+
# whether to bettertransformers
|
414 |
+
flash_optimum:
|
415 |
# whether to use xformers attention patch https://github.com/facebookresearch/xformers:
|
416 |
xformers_attention:
|
417 |
# whether to use flash attention patch https://github.com/HazyResearch/flash-attention:
|
scripts/finetune.py
CHANGED
@@ -14,7 +14,6 @@ import torch
|
|
14 |
import yaml
|
15 |
|
16 |
# add src to the pythonpath so we don't need to pip install this
|
17 |
-
from datasets import Dataset
|
18 |
from optimum.bettertransformer import BetterTransformer
|
19 |
from transformers import GenerationConfig, TextStreamer
|
20 |
|
@@ -208,14 +207,11 @@ def train(
|
|
208 |
tokenizer, cfg, DEFAULT_DATASET_PREPARED_PATH
|
209 |
)
|
210 |
else:
|
211 |
-
if cfg.pretraining_dataset is True:
|
212 |
-
pretraining_dataset = "togethercomputer/RedPajama-Data-1T"
|
213 |
-
else:
|
214 |
-
pretraining_dataset = cfg.pretraining_dataset
|
215 |
train_dataset = load_pretraining_dataset(
|
216 |
-
pretraining_dataset, tokenizer, max_tokens=cfg.sequence_len
|
217 |
)
|
218 |
-
|
|
|
219 |
eval_dataset = None
|
220 |
|
221 |
if cfg.debug or "debug" in kwargs:
|
@@ -262,19 +258,6 @@ def train(
|
|
262 |
model.save_pretrained(cfg.output_dir)
|
263 |
return
|
264 |
|
265 |
-
if cfg.debug:
|
266 |
-
logging.info("check_dataset_labels...")
|
267 |
-
check_dataset_labels(
|
268 |
-
train_dataset.select(
|
269 |
-
[random.randrange(0, len(train_dataset) - 1) for i in range(5)] # nosec
|
270 |
-
),
|
271 |
-
tokenizer,
|
272 |
-
)
|
273 |
-
|
274 |
-
if prepare_ds_only:
|
275 |
-
logging.info("Finished preparing dataset. Exiting...")
|
276 |
-
return
|
277 |
-
|
278 |
model.train()
|
279 |
|
280 |
trainer = setup_trainer(cfg, train_dataset, eval_dataset, model, tokenizer)
|
|
|
14 |
import yaml
|
15 |
|
16 |
# add src to the pythonpath so we don't need to pip install this
|
|
|
17 |
from optimum.bettertransformer import BetterTransformer
|
18 |
from transformers import GenerationConfig, TextStreamer
|
19 |
|
|
|
207 |
tokenizer, cfg, DEFAULT_DATASET_PREPARED_PATH
|
208 |
)
|
209 |
else:
|
|
|
|
|
|
|
|
|
210 |
train_dataset = load_pretraining_dataset(
|
211 |
+
cfg.pretraining_dataset, tokenizer, max_tokens=cfg.sequence_len
|
212 |
)
|
213 |
+
# https://discuss.huggingface.co/t/how-to-use-huggingface-trainer-streaming-datasets-without-wrapping-it-with-torchdatas-iterablewrapper/25230
|
214 |
+
train_dataset = train_dataset.with_format("torch")
|
215 |
eval_dataset = None
|
216 |
|
217 |
if cfg.debug or "debug" in kwargs:
|
|
|
258 |
model.save_pretrained(cfg.output_dir)
|
259 |
return
|
260 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
261 |
model.train()
|
262 |
|
263 |
trainer = setup_trainer(cfg, train_dataset, eval_dataset, model, tokenizer)
|
src/axolotl/utils/data.py
CHANGED
@@ -1,12 +1,12 @@
|
|
1 |
"""Module containing data utilities"""
|
2 |
-
|
3 |
import logging
|
4 |
from hashlib import md5
|
5 |
from pathlib import Path
|
6 |
from typing import List, Tuple, Union
|
7 |
|
8 |
import torch
|
9 |
-
from datasets import Dataset, DatasetDict,
|
10 |
from huggingface_hub import hf_hub_download
|
11 |
from transformers import PreTrainedTokenizerBase
|
12 |
|
@@ -399,32 +399,116 @@ def load_prepare_datasets(
|
|
399 |
return train_dataset, eval_dataset
|
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 |
def load_pretraining_dataset(path, tokenizer, max_tokens=2048):
|
430 |
-
|
|
|
|
|
|
|
|
|
|
|
|
1 |
"""Module containing data utilities"""
|
2 |
+
import functools
|
3 |
import logging
|
4 |
from hashlib import md5
|
5 |
from pathlib import Path
|
6 |
from typing import List, Tuple, Union
|
7 |
|
8 |
import torch
|
9 |
+
from datasets import Dataset, DatasetDict, load_dataset, load_from_disk
|
10 |
from huggingface_hub import hf_hub_download
|
11 |
from transformers import PreTrainedTokenizerBase
|
12 |
|
|
|
399 |
return train_dataset, eval_dataset
|
400 |
|
401 |
|
402 |
+
def encode_pretraining(tokenizer, max_tokens, examples):
|
403 |
+
res = tokenizer(
|
404 |
+
examples["text"],
|
405 |
+
truncation=True,
|
406 |
+
max_length=max_tokens - 2,
|
407 |
+
add_special_tokens=True,
|
408 |
+
)
|
409 |
+
# Convert to PyTorch tensors
|
410 |
+
input_ids = [torch.tensor(seq) for seq in res["input_ids"]]
|
411 |
+
attention_mask = [torch.tensor(seq) for seq in res["attention_mask"]]
|
412 |
+
new_input_ids = []
|
413 |
+
new_attention_mask = []
|
414 |
+
# Append EOS and PAD tokens to input_ids, and correct attention_mask
|
415 |
+
for i, _ in enumerate(input_ids):
|
416 |
+
input_ids[i] = torch.cat(
|
417 |
+
(
|
418 |
+
input_ids[i],
|
419 |
+
torch.tensor([tokenizer.eos_token_id, tokenizer.pad_token_id]),
|
420 |
+
),
|
421 |
+
dim=0,
|
422 |
+
)
|
423 |
+
attention_mask[i] = torch.cat((attention_mask[i], torch.tensor([1, 0])), dim=0)
|
424 |
+
|
425 |
+
# Concatenate tokens so that their lengths are less than max_tokens
|
426 |
+
buffer_input_ids = torch.tensor([], dtype=torch.long)
|
427 |
+
buffer_attention_mask = torch.tensor([], dtype=torch.long)
|
428 |
+
|
429 |
+
for ids, mask in zip(input_ids, attention_mask):
|
430 |
+
if buffer_input_ids.numel() == max_tokens:
|
431 |
+
new_input_ids.append(buffer_input_ids)
|
432 |
+
new_attention_mask.append(buffer_attention_mask)
|
433 |
+
buffer_input_ids = torch.tensor([], dtype=torch.long)
|
434 |
+
buffer_attention_mask = torch.tensor([], dtype=torch.long)
|
435 |
+
buffer_input_ids = torch.cat((buffer_input_ids, ids), dim=0)
|
436 |
+
buffer_attention_mask = torch.cat((buffer_attention_mask, mask), dim=0)
|
437 |
+
elif buffer_input_ids.numel() + ids.numel() <= max_tokens:
|
438 |
+
buffer_input_ids = torch.cat((buffer_input_ids, ids), dim=0)
|
439 |
+
buffer_attention_mask = torch.cat((buffer_attention_mask, mask), dim=0)
|
440 |
+
else:
|
441 |
+
buffer_input_ids = torch.cat(
|
442 |
+
(
|
443 |
+
buffer_input_ids,
|
444 |
+
torch.full(
|
445 |
+
(max_tokens - buffer_input_ids.numel(),),
|
446 |
+
tokenizer.pad_token_id,
|
447 |
+
dtype=torch.long,
|
448 |
+
),
|
449 |
+
),
|
450 |
+
dim=0,
|
451 |
+
)
|
452 |
+
buffer_attention_mask = torch.cat(
|
453 |
+
(
|
454 |
+
buffer_attention_mask,
|
455 |
+
torch.full(
|
456 |
+
(max_tokens - buffer_attention_mask.numel(),),
|
457 |
+
0,
|
458 |
+
dtype=torch.long,
|
459 |
+
),
|
460 |
+
),
|
461 |
+
dim=0,
|
462 |
+
)
|
463 |
+
new_input_ids.append(buffer_input_ids)
|
464 |
+
new_attention_mask.append(buffer_attention_mask)
|
465 |
+
buffer_input_ids = torch.tensor([], dtype=torch.long)
|
466 |
+
buffer_attention_mask = torch.tensor([], dtype=torch.long)
|
467 |
+
|
468 |
+
buffer_input_ids = torch.cat((buffer_input_ids, ids), dim=0)
|
469 |
+
buffer_attention_mask = torch.cat((buffer_attention_mask, mask), dim=0)
|
470 |
+
|
471 |
+
if buffer_input_ids.numel() > 0: # for any leftover tokens
|
472 |
+
while buffer_input_ids.numel() < max_tokens: # make all sequences equal in size
|
473 |
+
buffer_input_ids = torch.cat(
|
474 |
+
(
|
475 |
+
buffer_input_ids,
|
476 |
+
torch.full(
|
477 |
+
(max_tokens - buffer_input_ids.numel(),),
|
478 |
+
tokenizer.pad_token_id,
|
479 |
+
dtype=torch.long,
|
480 |
+
),
|
481 |
+
),
|
482 |
+
dim=0,
|
483 |
+
)
|
484 |
+
buffer_attention_mask = torch.cat(
|
485 |
+
(
|
486 |
+
buffer_attention_mask,
|
487 |
+
torch.full(
|
488 |
+
(max_tokens - buffer_attention_mask.numel(),),
|
489 |
+
0,
|
490 |
+
dtype=torch.long,
|
491 |
+
),
|
492 |
+
),
|
493 |
+
dim=0,
|
494 |
+
)
|
495 |
+
new_input_ids.append(buffer_input_ids)
|
496 |
+
new_attention_mask.append(buffer_attention_mask)
|
497 |
+
|
498 |
+
ret = {
|
499 |
+
"input_ids": [seq.tolist() for seq in new_input_ids],
|
500 |
+
"labels": [seq.tolist() for seq in new_input_ids],
|
501 |
+
"attention_mask": [seq.tolist() for seq in new_attention_mask],
|
502 |
+
}
|
503 |
+
|
504 |
+
logging.debug(len(ret["input_ids"]))
|
505 |
+
return ret
|
506 |
|
507 |
|
508 |
def load_pretraining_dataset(path, tokenizer, max_tokens=2048):
|
509 |
+
encode = functools.partial(encode_pretraining, tokenizer, max_tokens)
|
510 |
+
dataset = load_dataset(path, streaming=True, split="train")
|
511 |
+
dataset = dataset.shuffle(seed=42, buffer_size=10_000)
|
512 |
+
# TODO dynamically figure out which columns/features to remove
|
513 |
+
dataset = dataset.map(encode, batched=True, remove_columns=["text", "meta"])
|
514 |
+
return dataset
|
src/axolotl/utils/validation.py
CHANGED
@@ -77,6 +77,11 @@ def validate_config(cfg):
|
|
77 |
f"flash_optimum for BetterTransformers may not be used with {torch.__version__}"
|
78 |
)
|
79 |
|
|
|
|
|
|
|
|
|
|
|
80 |
# TODO
|
81 |
# MPT 7b
|
82 |
# https://github.com/facebookresearch/bitsandbytes/issues/25
|
|
|
77 |
f"flash_optimum for BetterTransformers may not be used with {torch.__version__}"
|
78 |
)
|
79 |
|
80 |
+
if cfg.pretraining_dataset and cfg.group_by_length:
|
81 |
+
logging.warning(
|
82 |
+
"You probably want to disable group_by_length as it will force a streamed dataset to download completely."
|
83 |
+
)
|
84 |
+
|
85 |
# TODO
|
86 |
# MPT 7b
|
87 |
# https://github.com/facebookresearch/bitsandbytes/issues/25
|
tests/test_validation.py
CHANGED
@@ -198,3 +198,54 @@ class ValidationTest(unittest.TestCase):
|
|
198 |
)
|
199 |
|
200 |
validate_config(cfg)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
198 |
)
|
199 |
|
200 |
validate_config(cfg)
|
201 |
+
|
202 |
+
def test_flash_optimum(self):
|
203 |
+
cfg = DictDefault(
|
204 |
+
{
|
205 |
+
"flash_optimum": True,
|
206 |
+
"adapter": "lora",
|
207 |
+
}
|
208 |
+
)
|
209 |
+
|
210 |
+
with self._caplog.at_level(logging.WARNING):
|
211 |
+
validate_config(cfg)
|
212 |
+
assert any(
|
213 |
+
"BetterTransformers probably doesn't work with PEFT adapters"
|
214 |
+
in record.message
|
215 |
+
for record in self._caplog.records
|
216 |
+
)
|
217 |
+
|
218 |
+
cfg = DictDefault(
|
219 |
+
{
|
220 |
+
"flash_optimum": True,
|
221 |
+
}
|
222 |
+
)
|
223 |
+
|
224 |
+
with self._caplog.at_level(logging.WARNING):
|
225 |
+
validate_config(cfg)
|
226 |
+
assert any(
|
227 |
+
"probably set bfloat16 or float16" in record.message
|
228 |
+
for record in self._caplog.records
|
229 |
+
)
|
230 |
+
|
231 |
+
cfg = DictDefault(
|
232 |
+
{
|
233 |
+
"flash_optimum": True,
|
234 |
+
"fp16": True,
|
235 |
+
}
|
236 |
+
)
|
237 |
+
regex_exp = r".*AMP is not supported.*"
|
238 |
+
|
239 |
+
with pytest.raises(ValueError, match=regex_exp):
|
240 |
+
validate_config(cfg)
|
241 |
+
|
242 |
+
cfg = DictDefault(
|
243 |
+
{
|
244 |
+
"flash_optimum": True,
|
245 |
+
"bf16": True,
|
246 |
+
}
|
247 |
+
)
|
248 |
+
regex_exp = r".*AMP is not supported.*"
|
249 |
+
|
250 |
+
with pytest.raises(ValueError, match=regex_exp):
|
251 |
+
validate_config(cfg)
|