vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

fix flake8 warnings

Browse source
This commit is contained in:
wonjun Jang 2023-11-28 16:46:54 +09:00 • committed by GitHub
parent 61edd1bc59
commit 1f5357cbcf
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 17 additions and 21 deletions
Download patch file Download diff file Expand all files Collapse all files

26
convert.py
View file

@ -22,7 +22,7 @@ from collections import OrderedDict
from concurrent.futures import ProcessPoolExecutor, ThreadPoolExecutor
from dataclasses import dataclass
from pathlib import Path
from typing import IO, TYPE_CHECKING, Any, Callable, Iterable, Literal, TypeVar
from typing import IO, TYPE_CHECKING, Any, Callable, Iterable, Literal, TypeVar, Optional
import numpy as np
from sentencepiece import SentencePieceProcessor
@ -310,10 +310,8 @@ class VocabLoader:
try:
self.tokenizer = AutoTokenizer.from_pretrained(str(fname_tokenizer), trust_remote_code=True)
vocab_set = {encoded_tok for encoded_tok, id in self.tokenizer.get_vocab().items()}
except:
except Exception:
self.tokenizer = AutoTokenizer.from_pretrained(str(fname_tokenizer), use_fast=False, trust_remote_code=True)
vocab_set = {encoded_tok for encoded_tok, id in self.tokenizer.get_vocab().items()}
self.added_tokens_dict: OrderedDict[str, int] = OrderedDict()
@ -423,15 +421,15 @@ class VocabLoader:
return "llama"
path_candidates.append(path_candidate)
raise FileNotFoundError(
f"Could not find {find_candidates} in {path} or its parent; "
"if it's in another directory, pass the directory as --vocab-dir")
raise FileNotFoundError(f"Could not find {path_candidates} in {self.fname_tokenizer} or its parent; if it's in another directory, pass the directory as --vocab-dir")
def __repr__(self) -> str:
return f"<VocabLoader with {self.vocab_size_base} base tokens and {len(self.added_tokens_dict)} added tokens>"
Vocab: TypeAlias = 'VocabLoader'
#
# data loading
# TODO: reuse (probably move to gguf.py?)
@ -806,6 +804,7 @@ def bounded_parallel_map(func: Callable[[In], Out], iterable: Iterable[In], conc
break
yield result
def check_vocab_size(params: Params, vocab: Vocab, pad_vocab: bool = False) -> None:
if params.n_vocab != vocab.vocab_size:
if params.n_vocab == vocab.vocab_size:
@ -907,11 +906,10 @@ class OutputFile:
self.gguf.close()
@staticmethod
def write_vocab_only(
fname_out: Path, params: Params, vocab: Vocab, svocab: gguf.SpecialVocab,
def write_vocab_only(fname_out: Path, params: Params, vocab: Vocab,
svocab: gguf.SpecialVocab,
endianess: gguf.GGUFEndian = gguf.GGUFEndian.LITTLE,
pad_vocab: bool = False,
) -> None:
pad_vocab: bool = False) -> None:
check_vocab_size(params, vocab, pad_vocab = pad_vocab)
of = OutputFile(fname_out, endianess=endianess)
@ -939,13 +937,11 @@ class OutputFile:
return dt.quantize(arr)
@staticmethod
def write_all(
fname_out : Path, ftype: GGMLFileType, params: Params,
def write_all(fname_out : Path, ftype: GGMLFileType, params: Params,
model : LazyModel, vocab: Vocab, svocab: gguf.SpecialVocab,
concurrency: int = DEFAULT_CONCURRENCY,
endianess : gguf.GGUFEndian = gguf.GGUFEndian.LITTLE,
pad_vocab : bool = False,
) -> None:
pad_vocab : bool = False) -> None:
check_vocab_size(params, vocab, pad_vocab = pad_vocab)
of = OutputFile(fname_out, endianess=endianess)