-
Notifications
You must be signed in to change notification settings - Fork 228
/
validate_data.py
372 lines (320 loc) · 13.6 KB
/
validate_data.py
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
import argparse
import json
from pathlib import Path
from typing import Dict
from mistral_common.exceptions import (
InvalidAssistantMessageException,
InvalidFunctionCallException,
InvalidMessageStructureException,
InvalidToolSchemaException,
TokenizerException,
)
from mistral_common.tokens.tokenizers.mistral import MistralTokenizer
from tqdm import tqdm
from finetune.args import TrainArgs
from finetune.data.dataset import parse_data_sources
from finetune.data.tokenize import (
ConversationFormatError,
FunctionFormatError,
MessageFormatError,
SampleType,
ToolCallFormatError,
UnrecognizedRoleError,
build_instruct_sample,
get_pretrain_sample,
tokenize,
)
NUM_GPUS = 8
# EXPECTED WPS for batch_size = 32768 per GPU on H100
EXPECTED_WPS = {
"open-mistral-7b": 5720,
"open-mixtral-8x7b": 2966,
"open-mixtral-8x22b": 1007,
"mistral-large-latest": 567,
'open-mistral-nemo': 3337,
}
MIN_NUM_JSONL_LINES = 10
MAX_NUM_JSONL_LINES = 10_000_000
MIN_BYTES = 1_000
MAX_BYTES = 10_000_000_000 # roughly 10 GB
def convert_seconds_to_hms(seconds: float) -> str:
hours = int(seconds // 3600)
seconds %= 3600
minutes = int(seconds // 60)
seconds %= 60
seconds = int(seconds)
return f"{hours:02d}:{minutes:02d}:{seconds:02d}"
def verify_size(jsonl_file: Path):
print(f"Verifying {jsonl_file} ...")
with jsonl_file.open() as f:
num_lines = 0
num_bytes = 0
for line in f:
num_lines += 1
num_bytes += len(line)
if num_lines < MIN_NUM_JSONL_LINES:
raise ValueError(
f"{jsonl_file} has only {num_lines} conversation which is less than the minimum amount of conversations required per dataset file: {MIN_NUM_JSONL_LINES}. Please make sure that each dataset has at least {MIN_NUM_JSONL_LINES} conversations."
)
elif num_bytes < MIN_BYTES:
raise ValueError(
f"{jsonl_file} has only {num_bytes} bytes which is less than the minimum amount of bytes required per dataset file: {MIN_BYTES}. Please make sure that each dataset has at least {MIN_BYTES} bytes."
)
elif num_lines > MAX_NUM_JSONL_LINES:
raise ValueError(
f"{jsonl_file} has {num_lines} conversations which is more than the maximum amount of allowed per dataset file: {MAX_NUM_JSONL_LINES}. Please make sure that each dataset has no more than {MAX_NUM_JSONL_LINES} conversations."
)
elif num_bytes > MAX_BYTES:
raise ValueError(
f"{jsonl_file} has {num_bytes} bytes which is more than the maximum amount of bytes allowed per dataset file: {MAX_BYTES}. Please make sure that each dataset has no more than {MAX_BYTES} bytes."
)
print(
f"Dataset {jsonl_file} is valid. Dataset has {num_lines} conversations amounting to a total of {num_bytes} bytes."
)
def get_train_stats(
num_tokens: Dict[str, int],
datasets_proportion: Dict[str, float],
train_args: TrainArgs,
return_type: str,
):
dataset_tokens = sum(num_tokens.values())
batch_size = train_args.batch_size * train_args.seq_len * NUM_GPUS
if Path(train_args.model_id_or_path).is_dir():
params_config = json.load(
(Path(train_args.model_id_or_path) / "params.json").open()
)
if params_config["dim"] == 4096 and params_config.get("moe") is None:
model_id = "open-mistral-7b"
elif params_config["dim"] == 4096 and params_config.get("moe") is not None:
model_id = "open-mixtral-8x7b"
elif params_config["dim"] == 6144:
model_id = "open-mixtral-8x22b"
elif params_config["dim"] == 12288:
model_id = "mistral-large-latest"
elif params_config["dim"] == 5120:
model_id = "open-mistral-nemo"
else:
raise ValueError("Provided model folder seems incorrect.")
else:
model_id = train_args.model_id_or_path
wps = EXPECTED_WPS[model_id]
if return_type == "expected":
train_tokens = train_args.max_steps * batch_size
max_steps = train_args.max_steps
num_epochs = train_tokens / dataset_tokens
elif return_type == "recommended":
num_epochs = 3
max_steps = int(sum(num_tokens.values()) / batch_size * num_epochs)
train_tokens = max_steps * batch_size
else:
raise ValueError(
f"`return_type` is {return_type}, but has to be one of ['expected', 'recommended']"
)
expected_time_in_sec = train_tokens / NUM_GPUS / wps
# Add 5min buffer for loading/init/ckpt/eval
expected_time_in_sec += 300
train_tokens_per_dataset = {
k: (train_tokens * v) for k, v in datasets_proportion.items()
}
return {
"eta": convert_seconds_to_hms(expected_time_in_sec),
"data_tokens": dataset_tokens,
"train_tokens": train_tokens,
"epochs": f"{num_epochs:.2f}",
"max_steps": max_steps,
"data_tokens_per_dataset": {k: f"{v:.1f}" for k, v in num_tokens.items()},
"train_tokens_per_dataset": {
k: f"{v:.1f}" for k, v in train_tokens_per_dataset.items()
},
"epochs_per_dataset": {
k: f"{(train_tokens_per_dataset[k] / num_tokens[k]):.1f}"
for k in num_tokens.keys()
},
}
def main(args):
train_args = TrainArgs.load(args.train_yaml)
yaml_data_errors = []
conversation_format_errors = []
message_format_errors = []
tokenization_errors = []
# Check if pretrain can be loaded
# train_pretrain_data = train_args.data.data
data = [("train", train_args.data.data, train_args.data.instruct_data)]
if train_args.data.eval_instruct_data != "":
data.append(("eval", "", train_args.data.eval_instruct_data))
EXPECTED_WPS.keys()
instruct_tokenizer = MistralTokenizer.v3().instruct_tokenizer
for name, pretrain_file, instruct_file in data:
datasets, weights = parse_data_sources(pretrain_file, instruct_file)
data_types = [d.sample_type for d in datasets]
datasets = [str(d.path) for d in datasets]
datasets_proportion = dict(zip(datasets, weights))
num_tokens = {k: 0 for k in datasets_proportion}
for data_type, dataset in tqdm(zip(data_types, datasets)):
# verify_size(Path(dataset))
print(f"Validating {dataset} ...")
corrected_dataset = dataset + ".corrected"
correct_lines = []
sub_yaml_data_errors = []
sub_conversation_format_errors = []
sub_message_format_errors = []
sub_tokenization_errors = []
# Load the dataset
with open(dataset, "r", encoding="utf-8") as f:
lines = f.readlines()
for idx, line in tqdm(enumerate(lines), total=len(lines)):
try:
data = json.loads(line)
except ValueError as e:
prefix = f"The data in line {idx + 1} of dataset {dataset} is incorrectly formatted."
sub_yaml_data_errors.append(prefix + str(e))
continue
if data_type == SampleType.PRETRAIN:
# TODO(Patrick) - Get good error message
sample = get_pretrain_sample(data)
else:
try:
sample = build_instruct_sample(data)
except (
ConversationFormatError,
UnrecognizedRoleError,
MessageFormatError,
ToolCallFormatError,
FunctionFormatError,
InvalidAssistantMessageException,
InvalidFunctionCallException,
InvalidMessageStructureException,
InvalidToolSchemaException,
) as e:
prefix = f"The data in line {idx + 1} of dataset {dataset} is incorrectly formatted."
if isinstance(
e, (ConversationFormatError, FunctionFormatError)
):
sub_conversation_format_errors.append(prefix + str(e))
elif isinstance(
e,
(
MessageFormatError,
UnrecognizedRoleError,
ToolCallFormatError,
),
):
sub_message_format_errors.append(prefix + str(e))
if isinstance(
e,
(
InvalidFunctionCallException,
InvalidMessageStructureException,
InvalidAssistantMessageException,
InvalidToolSchemaException,
),
):
sub_conversation_format_errors.append(prefix + str(e))
continue
try:
tokens = tokenize(sample, instruct_tokenizer).tokens
except TokenizerException as e:
error_message = (
f"The data in line {idx + 1} of dataset {dataset} could not be tokenized. "
+ str(e)
)
sub_tokenization_errors.append(error_message)
correct_lines.append(line)
num_tokens[dataset] += len(tokens)
is_sub_error = (
len(
sub_yaml_data_errors
+ sub_conversation_format_errors
+ sub_message_format_errors
+ sub_tokenization_errors
)
> 0
)
if is_sub_error and args.create_corrected:
with open(corrected_dataset, "w", encoding="utf-8") as f:
for line in correct_lines:
f.write(line)
print(f"Saved {corrected_dataset}.")
elif args.create_corrected:
print(f"No error in {dataset} - no need to create a corrected version.")
yaml_data_errors.extend(sub_yaml_data_errors)
conversation_format_errors.extend(sub_conversation_format_errors)
message_format_errors.extend(sub_message_format_errors)
tokenization_errors.extend(sub_tokenization_errors)
is_error = (
len(
yaml_data_errors
+ conversation_format_errors
+ message_format_errors
+ tokenization_errors
)
> 0
)
if is_error:
all_yaml_data_errors = "\n".join(yaml_data_errors)
all_conversation_format_errors = "\n".join(conversation_format_errors)
all_message_format_errors = "\n".join(message_format_errors)
all_tokenization_errors = "\n".join(tokenization_errors)
error_report = f"""
Data error report
----------------------- \n
The passed datasets contains some errors as listed below. Please make sure to fix these errors in order to start training.
YAML data load errors: \n\n {all_yaml_data_errors} \n\n
Conversation format errors: \n\n {all_conversation_format_errors} \n\n
Message format errors: \n\n {all_message_format_errors} \n\n
Tokenization errors: \n\n {all_tokenization_errors} \n\n
"""
if args.save_reports:
with open(args.error_report_txt, "w") as f:
f.write(error_report)
print(error_report)
else:
print("No errors! Data is correctly formatted!")
if name == "train" and not is_error:
expected_stats = get_train_stats(
num_tokens, datasets_proportion, train_args, return_type="expected"
)
stats = {
"expected": expected_stats,
}
filenames = (
f"{instruct_file}"
if pretrain_file == ""
else f"{instruct_file} and {pretrain_file}"
)
print(
f"Stats for {filenames} \n {20 * '-'} \n {json.dumps(stats, indent=4)}"
)
if args.save_reports:
if name == "train":
with open(args.train_stats_json, "w") as file:
json.dump(stats, file, indent=4)
if __name__ == "__main__":
parser = argparse.ArgumentParser(description="Validate jsonl.")
parser.add_argument(
"--train_yaml",
type=str,
help="Path to the data file",
)
parser.add_argument(
"--error_report_txt",
type=str,
default="data_errors.txt",
help="Path to the error report.",
)
parser.add_argument(
"--train_stats_json",
type=str,
default="train_stats.json",
help="Path to training statistics json file.",
)
parser.add_argument(
"--save_reports", action="store_true", help="Save reports to disk"
)
parser.add_argument(
"--create_corrected",
action="store_true",
help="Skip faulty lines and append all correct lines to `.corrected` datasets.",
)
args = parser.parse_args()
main(args)