From 34b0a082074b073eb14c2bd93c0c070e20ddcd16 Mon Sep 17 00:00:00 2001 From: Kerfuffle <44031344+KerfuffleV2@users.noreply.github.com> Date: Fri, 10 Nov 2023 22:04:50 -0700 Subject: gguf-py: Refactor and allow reading/modifying existing GGUF files (#3981) * gguf-py: Refactor and add file reading support * Replay changes from #3871 Credit to @cebtenzzre for that pull * Various type annotation fixes. * sort imports with isort (again) * Fix missing return statement in add_tensor * style cleanup with flake8 * fix NamedTuple and Enum usage * Fix an issue with state init in GGUFReader Move examples to an examples/ directory Clean up examples Add an example of modifying keys in a GGUF file Update documentation with info on examples Try to support people importing gguf/gguf.py directly * Damagage is not a word. * Clean up gguf-py/examples/modify_gguf.py whitespace Co-authored-by: Jared Van Bortel * Update gguf-py/examples/modify_gguf.py formatting Co-authored-by: Jared Van Bortel * Update gguf-py/gguf/gguf_reader.py type hint Co-authored-by: Jared Van Bortel * Make examples executable, formatting changes * Add more information to GGUFReader and examples comments * Include a gguf Python package version bump * Add convert-gguf-endian.py script * cleanup * gguf-py : bump minor version * Reorganize scripts * Make GGUFReader endian detection less arbitrary * Add JSON dumping support to gguf-dump.py Which I kind of regret now * A few for gguf-dump.py cleanups * Murder accidental tuple in gguf-py/scripts/gguf-dump.py Co-authored-by: Jared Van Bortel * cleanup * constants : remove unneeded type annotations * fix python 3.8 compat * Set up gguf- scripts in pyproject.toml * And include scripts/__init__.py, derp * convert.py: We can't currently support Q8_0 on big endian. * gguf-py: SpecialVocab: Always try available sources for special token ids gguf-py: SpecialVocab: Try to load merges from merges.txt if not in tokenizer.json gguf-py: SpecialVocab: Add 'add_bos_token' type bools to GGUF metadata u * cleanup * Promote add_X_token to GGUF metadata for BOS and EOS --------- Co-authored-by: Jared Van Bortel Co-authored-by: Jared Van Bortel --- convert.py | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) (limited to 'convert.py') diff --git a/convert.py b/convert.py index b0f44dbe..a4b87e08 100755 --- a/convert.py +++ b/convert.py @@ -3,11 +3,9 @@ from __future__ import annotations import argparse import concurrent.futures -import copy import enum import faulthandler import functools -import io import itertools import json import math @@ -23,14 +21,14 @@ from abc import ABCMeta, abstractmethod from concurrent.futures import ProcessPoolExecutor, ThreadPoolExecutor from dataclasses import dataclass from pathlib import Path -from typing import IO, TYPE_CHECKING, Any, Callable, Generator, Iterable, Literal, Sequence, TypeVar +from typing import IO, TYPE_CHECKING, Any, Callable, Iterable, Literal, TypeVar import numpy as np from sentencepiece import SentencePieceProcessor import os if 'NO_LOCAL_GGUF' not in os.environ: - sys.path.insert(1, str(Path(__file__).parent / 'gguf-py' / 'gguf')) + sys.path.insert(1, str(Path(__file__).parent / 'gguf-py')) import gguf if TYPE_CHECKING: @@ -851,7 +849,7 @@ class OutputFile: elif isinstance(vocab, BpeVocab): self.gguf.add_tokenizer_model("gpt2") else: - raise ValueError(f'Unknown vocab type: Not BpeVocab or SentencePieceVocab') + raise ValueError('Unknown vocab type: Not BpeVocab or SentencePieceVocab') self.gguf.add_token_list(tokens) self.gguf.add_token_scores(scores) self.gguf.add_token_types(toktypes) @@ -905,7 +903,7 @@ class OutputFile: return dt.quantize(arr) @staticmethod - def write_all(fname_out: Path, ftype: GGMLFileType, params: Params, model: LazyModel, vocab: Vocab, svocab: gguf.SpecialVocab, concurrency: int = DEFAULT_CONCURRENCY, endianess=gguf.GGUFEndian.LITTLE) -> None: + 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) -> None: check_vocab_size(params, vocab) of = OutputFile(fname_out, endianess=endianess) @@ -1114,11 +1112,15 @@ def do_dump_model(model_plus: ModelPlus) -> None: def main(args_in: list[str] | None = None) -> None: + output_choices = ["f32", "f16"] + if np.uint32(1) == np.uint32(1).newbyteorder("<"): + # We currently only support Q8_0 output on little endian systems. + output_choices.append("q8_0") parser = argparse.ArgumentParser(description="Convert a LLaMa model to a GGML compatible file") parser.add_argument("--dump", action="store_true", help="don't convert, just show what's in the model") parser.add_argument("--dump-single", action="store_true", help="don't convert, just show what's in a single model file") parser.add_argument("--vocab-only", action="store_true", help="extract only the vocab") - parser.add_argument("--outtype", choices=["f32", "f16", "q8_0"], help="output format - note: q8_0 may be very slow (default: f16 or f32 based on input)") + parser.add_argument("--outtype", choices=output_choices, help="output format - note: q8_0 may be very slow (default: f16 or f32 based on input)") parser.add_argument("--vocab-dir", type=Path, help="directory containing tokenizer.model, if separate from model file") parser.add_argument("--outfile", type=Path, help="path to write to; default: based on input") parser.add_argument("model", type=Path, help="directory containing model file, or model file itself (*.pth, *.pt, *.bin)") -- cgit v1.2.3