-
Notifications
You must be signed in to change notification settings - Fork 38
/
hf_api_endpoint.py
255 lines (228 loc) · 8.33 KB
/
hf_api_endpoint.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
import gc
import logging
import re
import urllib.parse
from concurrent.futures import ThreadPoolExecutor
from functools import cached_property, partial
from typing import Any, Callable, Generator, Iterable
from datasets.fingerprint import Hasher
from tenacity import (
after_log,
before_sleep_log,
retry,
retry_if_exception_type,
stop_any,
wait_exponential,
)
from .._cachable._cachable import _StrWithSeed
from ..utils.arg_utils import AUTO, Default
from ..utils.fs_utils import safe_fn
from ..utils.import_utils import ignore_pydantic_warnings
from .hf_transformers import CachedTokenizer, HFTransformers
from .llm import (
DEFAULT_BATCH_SIZE,
LLM,
_check_max_new_tokens_possible,
_check_temperature_and_top_p,
)
with ignore_pydantic_warnings():
from huggingface_hub import InferenceClient
class HFAPIEndpoint(HFTransformers):
def __init__(
self,
endpoint: str,
model_name: str,
chat_prompt_template: None | str | Default = AUTO,
system_prompt: None | str | Default = AUTO,
token: None | str = None,
revision: None | str = None,
trust_remote_code: bool = False,
retry_on_fail: bool = True,
cache_folder_path: None | str = None,
**kwargs,
):
super().__init__(
model_name=model_name,
chat_prompt_template=chat_prompt_template,
system_prompt=system_prompt,
revision=revision,
trust_remote_code=trust_remote_code,
cache_folder_path=cache_folder_path,
**kwargs,
)
self.endpoint = endpoint
self.token = token
# Setup API calling helpers
self.retry_on_fail = retry_on_fail
self.executor_pools: dict[int, ThreadPoolExecutor] = {}
@cached_property
def retry_wrapper(self):
# Create a retry wrapper function
tenacity_logger = self.get_logger(key="retry", verbose=True, log_level=None)
@retry(
retry=retry_if_exception_type(Exception),
wait=wait_exponential(multiplier=1, min=3, max=300),
before_sleep=before_sleep_log(tenacity_logger, logging.INFO),
after=after_log(tenacity_logger, logging.INFO),
stop=stop_any(lambda _: not self.retry_on_fail), # type: ignore[arg-type]
reraise=True,
)
def _retry_wrapper(func, **kwargs):
return func(**kwargs)
_retry_wrapper.__wrapped__.__module__ = None # type: ignore[attr-defined]
_retry_wrapper.__wrapped__.__qualname__ = f"{self.__class__.__name__}.run" # type: ignore[attr-defined]
return _retry_wrapper
@cached_property
def client(self) -> InferenceClient:
return InferenceClient(model=self.endpoint, token=self.token, **self.kwargs)
def _is_batch_size_exception(self, e: BaseException) -> bool: # pragma: no cover
return False
def _run_batch(
self,
cached_tokenizer: CachedTokenizer,
max_length_func: Callable[[list[str]], int],
inputs: list[str],
max_new_tokens: None | int = None,
temperature: float = 1.0,
top_p: float = 0.0,
n: int = 1,
stop: None | str | list[str] = None,
repetition_penalty: None | float = None,
logit_bias: None | dict[int, float] = None,
batch_size: int = DEFAULT_BATCH_SIZE,
seed: None | int = None,
**kwargs,
) -> list[str] | list[list[str]]:
prompts = inputs
assert (
logit_bias is None
), f"`logit_bias` is not supported for {type(self).__name__}"
assert n == 1, f"Only `n` = 1 is supported for {type(self).__name__}"
# Check max_new_tokens
max_new_tokens = _check_max_new_tokens_possible(
self=self,
max_length_func=max_length_func,
prompts=prompts,
max_new_tokens=max_new_tokens,
)
# Set temperature and top_p
temperature, top_p = _check_temperature_and_top_p(
temperature=temperature,
top_p=top_p,
supports_zero_temperature=False,
supports_zero_top_p=False,
supports_one_top_p=False,
)
# Run the model
def get_generated_texts(self, kwargs, prompt) -> list[str]:
generated_text = self.retry_wrapper(
func=self.client.text_generation,
model=self.endpoint,
prompt=prompt,
do_sample=True,
max_new_tokens=max_new_tokens,
repetition_penalty=repetition_penalty,
return_full_text=False,
seed=(
seed + _StrWithSeed.total_per_input_seeds([prompt])
if seed is not None
else None
),
stop_sequences=stop,
temperature=temperature,
top_p=top_p,
**kwargs,
)
return [generated_text]
if batch_size not in self.executor_pools:
self.executor_pools[batch_size] = ThreadPoolExecutor(max_workers=batch_size)
generated_texts_batch = list(
self.executor_pools[batch_size].map(
partial(get_generated_texts, self, kwargs), prompts
)
)
if n == 1:
return [batch[0] for batch in generated_texts_batch]
else: # pragma: no cover
return generated_texts_batch
def run(
self,
prompts: Iterable[str],
max_new_tokens: None | int = None,
temperature: float = 1.0,
top_p: float = 0.0,
n: int = 1,
stop: None | str | list[str] = None,
repetition_penalty: None | float = None,
logit_bias: None | dict[int, float] = None,
batch_size: int = DEFAULT_BATCH_SIZE,
batch_scheduler_buffer_size: None | int = None,
adaptive_batch_size: bool = False,
seed: None | int = None,
progress_interval: None | int = 60,
force: bool = False,
cache_only: bool = False,
verbose: None | bool = None,
log_level: None | int = None,
total_num_prompts: None | int = None,
return_generator: bool = False,
**kwargs,
) -> Generator[str | list[str], None, None] | list[str | list[str]]:
return super().run(
prompts=prompts,
max_new_tokens=max_new_tokens,
temperature=temperature,
top_p=top_p,
n=n,
stop=stop,
repetition_penalty=repetition_penalty,
logit_bias=logit_bias,
batch_size=batch_size,
batch_scheduler_buffer_size=batch_scheduler_buffer_size,
adaptive_batch_size=adaptive_batch_size,
seed=seed,
progress_interval=progress_interval,
force=force,
cache_only=cache_only,
verbose=verbose,
log_level=log_level,
total_num_prompts=total_num_prompts,
return_generator=return_generator,
**kwargs,
)
@cached_property
def display_name(self) -> str:
name = (
re.sub(r"/(.*)", r"\1", urllib.parse.urlparse(self.endpoint).path).strip()
or self.endpoint
)
return LLM.display_name.func(self) + f" ({name})" # type: ignore[attr-defined]
@cached_property
def _cache_name(self) -> None | str:
names = [
safe_fn(self.endpoint, allow_slashes=False),
safe_fn(self.model_name, allow_slashes=False),
]
if self.revision:
names.append(self.revision)
to_hash: list[Any] = []
if len(to_hash) > 0: # pragma: no cover
names.append(Hasher.hash(to_hash))
return "_".join(names)
def unload_model(self):
# Delete cached client and tokenizer
if "client" in self.__dict__:
del self.__dict__["client"]
if "tokenizer" in self.__dict__:
del self.__dict__["tokenizer"]
# Garbage collect
gc.collect()
def __getstate__(self): # pragma: no cover
state = super().__getstate__()
# Remove cached client or tokenizer before serializing
state.pop("retry_wrapper", None)
state.pop("client", None)
state.pop("tokenizer", None)
state["executor_pools"].clear()
return state
__all__ = ["HFAPIEndpoint"]