Update deprecated Python 3.8 typing (#13971)
This commit is contained in:
@@ -1,6 +1,7 @@
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
|
||||
from typing import Any, Dict, Generator, List, Optional
|
||||
from collections.abc import Generator
|
||||
from typing import Any, Optional
|
||||
|
||||
import pytest
|
||||
from transformers import AutoTokenizer
|
||||
@@ -163,7 +164,7 @@ def detokenizer(tokenizer_name: str) -> Detokenizer:
|
||||
|
||||
@pytest.fixture(name="complete_sequence_token_ids")
|
||||
def create_complete_sequence_token_ids(complete_sequence: str,
|
||||
tokenizer) -> List[int]:
|
||||
tokenizer) -> list[int]:
|
||||
complete_sequence_token_ids = tokenizer(complete_sequence).input_ids
|
||||
return complete_sequence_token_ids
|
||||
|
||||
@@ -178,7 +179,7 @@ def create_sequence(prompt_token_ids=None):
|
||||
|
||||
|
||||
def create_dummy_logprobs(
|
||||
complete_sequence_token_ids: List[int]) -> List[Dict[int, Logprob]]:
|
||||
complete_sequence_token_ids: list[int]) -> list[dict[int, Logprob]]:
|
||||
return [{
|
||||
token_id: Logprob(logprob=0.0),
|
||||
token_id + 1: Logprob(logprob=0.1)
|
||||
@@ -186,10 +187,10 @@ def create_dummy_logprobs(
|
||||
|
||||
|
||||
def create_dummy_prompt_logprobs(
|
||||
complete_sequence_token_ids: List[int]
|
||||
) -> List[Optional[Dict[int, Any]]]:
|
||||
complete_sequence_token_ids: list[int]
|
||||
) -> list[Optional[dict[int, Any]]]:
|
||||
# logprob for the first prompt token is None.
|
||||
logprobs: List[Optional[Dict[int, Any]]] = [None]
|
||||
logprobs: list[Optional[dict[int, Any]]] = [None]
|
||||
logprobs.extend(create_dummy_logprobs(complete_sequence_token_ids)[1:])
|
||||
return logprobs
|
||||
|
||||
@@ -198,7 +199,7 @@ def create_dummy_prompt_logprobs(
|
||||
@pytest.mark.parametrize("tokenizer_name", TOKENIZERS)
|
||||
@pytest.mark.parametrize("skip_special_tokens", [True, False], indirect=True)
|
||||
def test_decode_sequence_logprobs(complete_sequence: str,
|
||||
complete_sequence_token_ids: List[int],
|
||||
complete_sequence_token_ids: list[int],
|
||||
detokenizer: Detokenizer,
|
||||
skip_special_tokens: bool):
|
||||
"""Verify Detokenizer decodes logprobs correctly."""
|
||||
@@ -208,8 +209,8 @@ def test_decode_sequence_logprobs(complete_sequence: str,
|
||||
# Run sequentially.
|
||||
seq = create_sequence()
|
||||
dummy_logprobs = create_dummy_logprobs(complete_sequence_token_ids)
|
||||
sequential_logprobs_text_chosen_token: List[str] = []
|
||||
sequential_logprobs_text_other_token: List[str] = []
|
||||
sequential_logprobs_text_chosen_token: list[str] = []
|
||||
sequential_logprobs_text_other_token: list[str] = []
|
||||
for new_token, logprobs in zip(complete_sequence_token_ids,
|
||||
dummy_logprobs):
|
||||
seq.append_token_id(new_token, logprobs)
|
||||
@@ -232,7 +233,7 @@ def test_decode_sequence_logprobs(complete_sequence: str,
|
||||
|
||||
@pytest.mark.parametrize("complete_sequence", TRUTH)
|
||||
@pytest.mark.parametrize("tokenizer_name", TOKENIZERS)
|
||||
def test_decode_prompt_logprobs(complete_sequence_token_ids: List[int],
|
||||
def test_decode_prompt_logprobs(complete_sequence_token_ids: list[int],
|
||||
detokenizer: Detokenizer):
|
||||
"""Verify Detokenizer decodes prompt logprobs correctly."""
|
||||
sampling_params = SamplingParams(skip_special_tokens=True,
|
||||
@@ -249,7 +250,7 @@ def test_decode_prompt_logprobs(complete_sequence_token_ids: List[int],
|
||||
dummy_logprobs,
|
||||
position_offset=0)
|
||||
# First logprob is None.
|
||||
decoded_prompt_logprobs: List[Dict[int, Any]] = dummy_logprobs[
|
||||
decoded_prompt_logprobs: list[dict[int, Any]] = dummy_logprobs[
|
||||
1:] # type: ignore
|
||||
|
||||
# decoded_prompt_logprobs doesn't contain the first token.
|
||||
|
||||
@@ -3,7 +3,7 @@
|
||||
import asyncio
|
||||
import os
|
||||
import sys
|
||||
from typing import List, Optional
|
||||
from typing import Optional
|
||||
from unittest.mock import patch
|
||||
|
||||
import pytest
|
||||
@@ -129,7 +129,7 @@ async def test_tokenizer_group_ray_pool_fault_tolerance(tokenizer_group_type):
|
||||
|
||||
def __init__(self,
|
||||
*args,
|
||||
fail_at: Optional[List[int]] = None,
|
||||
fail_at: Optional[list[int]] = None,
|
||||
**kwargs):
|
||||
super().__init__(*args, **kwargs)
|
||||
self.i = 0
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# SPDX-License-Identifier: Apache-2.0
|
||||
|
||||
from typing import TYPE_CHECKING, Any, Dict, List, Optional, Union
|
||||
from typing import TYPE_CHECKING, Any, Optional, Union
|
||||
|
||||
from vllm.transformers_utils.tokenizer import get_tokenizer
|
||||
from vllm.transformers_utils.tokenizer_base import (TokenizerBase,
|
||||
@@ -17,15 +17,15 @@ class TestTokenizer(TokenizerBase):
|
||||
return TestTokenizer()
|
||||
|
||||
@property
|
||||
def all_special_tokens_extended(self) -> List[str]:
|
||||
def all_special_tokens_extended(self) -> list[str]:
|
||||
raise NotImplementedError()
|
||||
|
||||
@property
|
||||
def all_special_tokens(self) -> List[str]:
|
||||
def all_special_tokens(self) -> list[str]:
|
||||
raise NotImplementedError()
|
||||
|
||||
@property
|
||||
def all_special_ids(self) -> List[int]:
|
||||
def all_special_ids(self) -> list[int]:
|
||||
raise NotImplementedError()
|
||||
|
||||
@property
|
||||
@@ -58,7 +58,7 @@ class TestTokenizer(TokenizerBase):
|
||||
|
||||
def __call__(
|
||||
self,
|
||||
text: Union[str, List[str], List[int]],
|
||||
text: Union[str, list[str], list[int]],
|
||||
text_pair: Optional[str] = None,
|
||||
add_special_tokens: bool = False,
|
||||
truncation: bool = False,
|
||||
@@ -66,10 +66,10 @@ class TestTokenizer(TokenizerBase):
|
||||
):
|
||||
raise NotImplementedError()
|
||||
|
||||
def get_vocab(self) -> Dict[str, int]:
|
||||
def get_vocab(self) -> dict[str, int]:
|
||||
raise NotImplementedError()
|
||||
|
||||
def get_added_vocab(self) -> Dict[str, int]:
|
||||
def get_added_vocab(self) -> dict[str, int]:
|
||||
raise NotImplementedError()
|
||||
|
||||
def encode_one(
|
||||
@@ -77,33 +77,33 @@ class TestTokenizer(TokenizerBase):
|
||||
text: str,
|
||||
truncation: bool = False,
|
||||
max_length: Optional[int] = None,
|
||||
) -> List[int]:
|
||||
) -> list[int]:
|
||||
raise NotImplementedError()
|
||||
|
||||
def encode(self,
|
||||
text: str,
|
||||
add_special_tokens: Optional[bool] = None) -> List[int]:
|
||||
add_special_tokens: Optional[bool] = None) -> list[int]:
|
||||
raise NotImplementedError()
|
||||
|
||||
def apply_chat_template(self,
|
||||
messages: List["ChatCompletionMessageParam"],
|
||||
tools: Optional[List[Dict[str, Any]]] = None,
|
||||
**kwargs) -> List[int]:
|
||||
messages: list["ChatCompletionMessageParam"],
|
||||
tools: Optional[list[dict[str, Any]]] = None,
|
||||
**kwargs) -> list[int]:
|
||||
raise NotImplementedError()
|
||||
|
||||
def convert_tokens_to_string(self, tokens: List[str]) -> str:
|
||||
def convert_tokens_to_string(self, tokens: list[str]) -> str:
|
||||
raise NotImplementedError()
|
||||
|
||||
def decode(self,
|
||||
ids: Union[List[int], int],
|
||||
ids: Union[list[int], int],
|
||||
skip_special_tokens: bool = True) -> str:
|
||||
raise NotImplementedError()
|
||||
|
||||
def convert_ids_to_tokens(
|
||||
self,
|
||||
ids: List[int],
|
||||
ids: list[int],
|
||||
skip_special_tokens: bool = True,
|
||||
) -> List[str]:
|
||||
) -> list[str]:
|
||||
raise NotImplementedError()
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user