From eedad2af57ccacfc805c04e23ef7f36cabb51251 Mon Sep 17 00:00:00 2001 From: "sungho.park" Date: Wed, 15 Mar 2023 06:32:16 +0000 Subject: [PATCH] Fix training codes --- finetune.py | 6 +- lengths.ipynb | 118 ++++++++- requirements.txt | 4 +- train.py | 641 +++++++++++++++++++++++++++++++++++++++++++++++ train.sh | 8 + 5 files changed, 773 insertions(+), 4 deletions(-) create mode 100644 train.py create mode 100644 train.sh diff --git a/finetune.py b/finetune.py index 92566c8..0c269d1 100644 --- a/finetune.py +++ b/finetune.py @@ -20,11 +20,13 @@ LORA_ALPHA = 16 LORA_DROPOUT = 0.05 +model_path = os.environ["model_path"] + model = LLaMAForCausalLM.from_pretrained( - "decapoda-research/llama-7b-hf", + model_path, ) tokenizer = LLaMATokenizer.from_pretrained( - "decapoda-research/llama-7b-hf", add_eos_token=True + model_path, add_eos_token=True ) # model = prepare_model_for_int8_training(model) diff --git a/lengths.ipynb b/lengths.ipynb index e4a01ad..30ddaeb 100644 --- a/lengths.ipynb +++ b/lengths.ipynb @@ -1,5 +1,121 @@ { "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/conda/envs/alfh/lib/python3.10/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n", + " from .autonotebook import tqdm as notebook_tqdm\n", + "Found cached dataset json (/home/irteam/.cache/huggingface/datasets/json/default-801198b68acc55bc/0.0.0/0f7e3662623656454fcd2b650f34e886a7db4b9104504885bd462096cc7a9f51)\n", + "100%|██████████| 1/1 [00:00<00:00, 173.13it/s]\n" + ] + } + ], + "source": [ + "from datasets import load_dataset, Dataset, DatasetDict\n", + "data = load_dataset(\"json\", data_files=\"alpaca_data.json\")" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "DatasetDict({\n", + " train: Dataset({\n", + " features: ['instruction', 'input', 'output'],\n", + " num_rows: 52002\n", + " })\n", + "})" + ] + }, + "execution_count": 2, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "Dataset({\n", + " features: ['instruction', 'input', 'output'],\n", + " num_rows: 52002\n", + "})" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "data['train']" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [ + { + "ename": "KeyError", + "evalue": "'valid'", + "output_type": "error", + "traceback": [ + "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", + "\u001b[0;31mKeyError\u001b[0m Traceback (most recent call last)", + "Cell \u001b[0;32mIn[10], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m data[\u001b[39m'\u001b[39;49m\u001b[39mvalid\u001b[39;49m\u001b[39m'\u001b[39;49m]\n", + "File \u001b[0;32m/opt/conda/envs/alfh/lib/python3.10/site-packages/datasets/dataset_dict.py:58\u001b[0m, in \u001b[0;36mDatasetDict.__getitem__\u001b[0;34m(self, k)\u001b[0m\n\u001b[1;32m 56\u001b[0m \u001b[39mdef\u001b[39;00m \u001b[39m__getitem__\u001b[39m(\u001b[39mself\u001b[39m, k) \u001b[39m-\u001b[39m\u001b[39m>\u001b[39m Dataset:\n\u001b[1;32m 57\u001b[0m \u001b[39mif\u001b[39;00m \u001b[39misinstance\u001b[39m(k, (\u001b[39mstr\u001b[39m, NamedSplit)) \u001b[39mor\u001b[39;00m \u001b[39mlen\u001b[39m(\u001b[39mself\u001b[39m) \u001b[39m==\u001b[39m \u001b[39m0\u001b[39m:\n\u001b[0;32m---> 58\u001b[0m \u001b[39mreturn\u001b[39;00m \u001b[39msuper\u001b[39;49m()\u001b[39m.\u001b[39;49m\u001b[39m__getitem__\u001b[39;49m(k)\n\u001b[1;32m 59\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[1;32m 60\u001b[0m available_suggested_splits \u001b[39m=\u001b[39m [\n\u001b[1;32m 61\u001b[0m split \u001b[39mfor\u001b[39;00m split \u001b[39min\u001b[39;00m (Split\u001b[39m.\u001b[39mTRAIN, Split\u001b[39m.\u001b[39mTEST, Split\u001b[39m.\u001b[39mVALIDATION) \u001b[39mif\u001b[39;00m split \u001b[39min\u001b[39;00m \u001b[39mself\u001b[39m\n\u001b[1;32m 62\u001b[0m ]\n", + "\u001b[0;31mKeyError\u001b[0m: 'valid'" + ] + } + ], + "source": [ + "from sklearn.model_selection import train_test_split\n", + "\n", + "X_train, X_test, y_train, y_test = train_test_split(data['train']['caption'], data['tag'], test_size=0.2, random_state=42)\n", + "X_val, X_test, y_val, y_test = train_test_split(X_test, y_test, test_size=0.5, random_state=42)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "dataset = DatasetDict({\n", + " 'train': Dataset.from_dict({\n", + " 'caption': X_train,\n", + " 'tag': y_train\n", + " }),\n", + " 'validation': Dataset.from_dict({\n", + " 'caption': X_val,\n", + " 'tag': y_val\n", + " }),\n", + " 'test': Dataset.from_dict({\n", + " 'caption': X_test,\n", + " 'tag': y_test\n", + " })\n", + "})" + ] + }, { "cell_type": "code", "execution_count": 1, @@ -166,7 +282,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.10.8" + "version": "3.10.9" }, "orig_nbformat": 4, "vscode": { diff --git a/requirements.txt b/requirements.txt index b6fe2da..a3d17e2 100644 --- a/requirements.txt +++ b/requirements.txt @@ -5,4 +5,6 @@ loralib lora datasets peft -accelerate \ No newline at end of file +accelerate +evaluate +scikit-learn \ No newline at end of file diff --git a/train.py b/train.py new file mode 100644 index 0000000..1891f1c --- /dev/null +++ b/train.py @@ -0,0 +1,641 @@ +#!/usr/bin/env python +# coding=utf-8 +# Copyright 2020 The HuggingFace Inc. team. All rights reserved. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +""" +Fine-tuning the library models for causal language modeling (GPT, GPT-2, CTRL, ...) on a text file or a dataset. + +Here is the full list of checkpoints on the hub that can be fine-tuned by this script: +https://huggingface.co/models?filter=text-generation +""" +# You can also adapt this script on your own causal language modeling task. Pointers for this are left as comments. + +import logging +import math +import os +import sys +from dataclasses import dataclass, field +from itertools import chain +from typing import Optional + +import datasets +import evaluate +import torch +from datasets import load_dataset + +import transformers +from transformers import ( + CONFIG_MAPPING, + MODEL_FOR_CAUSAL_LM_MAPPING, + AutoConfig, + AutoModelForCausalLM, + AutoTokenizer, + LLaMAForCausalLM, LLaMATokenizer, + HfArgumentParser, + Trainer, + TrainingArguments, + default_data_collator, + is_torch_tpu_available, + set_seed, +) +from transformers.testing_utils import CaptureLogger +from transformers.trainer_utils import get_last_checkpoint +from transformers.utils import check_min_version, send_example_telemetry +from transformers.utils.versions import require_version + + +logger = logging.getLogger(__name__) + + +MODEL_CONFIG_CLASSES = list(MODEL_FOR_CAUSAL_LM_MAPPING.keys()) +MODEL_TYPES = tuple(conf.model_type for conf in MODEL_CONFIG_CLASSES) + + +@dataclass +class ModelArguments: + """ + Arguments pertaining to which model/config/tokenizer we are going to fine-tune, or train from scratch. + """ + + model_name_or_path: Optional[str] = field( + default=None, + metadata={ + "help": ( + "The model checkpoint for weights initialization.Don't set if you want to train a model from scratch." + ) + }, + ) + model_type: Optional[str] = field( + default=None, + metadata={"help": "If training from scratch, pass a model type from the list: " + ", ".join(MODEL_TYPES)}, + ) + config_overrides: Optional[str] = field( + default=None, + metadata={ + "help": ( + "Override some existing default config settings when a model is trained from scratch. Example: " + "n_embd=10,resid_pdrop=0.2,scale_attn_weights=false,summary_type=cls_index" + ) + }, + ) + config_name: Optional[str] = field( + default=None, metadata={"help": "Pretrained config name or path if not the same as model_name"} + ) + tokenizer_name: Optional[str] = field( + default=None, metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"} + ) + cache_dir: Optional[str] = field( + default=None, + metadata={"help": "Where do you want to store the pretrained models downloaded from huggingface.co"}, + ) + use_fast_tokenizer: bool = field( + default=True, + metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."}, + ) + model_revision: str = field( + default="main", + metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."}, + ) + use_auth_token: bool = field( + default=False, + metadata={ + "help": ( + "Will use the token generated when running `huggingface-cli login` (necessary to use this script " + "with private models)." + ) + }, + ) + torch_dtype: Optional[str] = field( + default=None, + metadata={ + "help": ( + "Override the default `torch.dtype` and load the model under this dtype. If `auto` is passed, the " + "dtype will be automatically derived from the model's weights." + ), + "choices": ["auto", "bfloat16", "float16", "float32"], + }, + ) + + def __post_init__(self): + if self.config_overrides is not None and (self.config_name is not None or self.model_name_or_path is not None): + raise ValueError( + "--config_overrides can't be used in combination with --config_name or --model_name_or_path" + ) + + +@dataclass +class DataTrainingArguments: + """ + Arguments pertaining to what data we are going to input our model for training and eval. + """ + + dataset_name: Optional[str] = field( + default=None, metadata={"help": "The name of the dataset to use (via the datasets library)."} + ) + dataset_config_name: Optional[str] = field( + default=None, metadata={"help": "The configuration name of the dataset to use (via the datasets library)."} + ) + train_file: Optional[str] = field(default=None, metadata={"help": "The input training data file (a text file)."}) + validation_file: Optional[str] = field( + default=None, + metadata={"help": "An optional input evaluation data file to evaluate the perplexity on (a text file)."}, + ) + max_train_samples: Optional[int] = field( + default=None, + metadata={ + "help": ( + "For debugging purposes or quicker training, truncate the number of training examples to this " + "value if set." + ) + }, + ) + max_eval_samples: Optional[int] = field( + default=None, + metadata={ + "help": ( + "For debugging purposes or quicker training, truncate the number of evaluation examples to this " + "value if set." + ) + }, + ) + streaming: bool = field(default=False, metadata={"help": "Enable streaming mode"}) + block_size: Optional[int] = field( + default=None, + metadata={ + "help": ( + "Optional input sequence length after tokenization. " + "The training dataset will be truncated in block of this size for training. " + "Default to the model max input length for single sentence inputs (take into account special tokens)." + ) + }, + ) + overwrite_cache: bool = field( + default=False, metadata={"help": "Overwrite the cached training and evaluation sets"} + ) + validation_split_percentage: Optional[int] = field( + default=5, + metadata={ + "help": "The percentage of the train set used as validation set in case there's no validation split" + }, + ) + preprocessing_num_workers: Optional[int] = field( + default=None, + metadata={"help": "The number of processes to use for the preprocessing."}, + ) + keep_linebreaks: bool = field( + default=True, metadata={"help": "Whether to keep line breaks when using TXT files or not."} + ) + + def __post_init__(self): + if self.streaming: + require_version("datasets>=2.0.0", "The streaming feature requires `datasets>=2.0.0`") + + if self.dataset_name is None and self.train_file is None and self.validation_file is None: + raise ValueError("Need either a dataset name or a training/validation file.") + else: + if self.train_file is not None: + extension = self.train_file.split(".")[-1] + assert extension in ["csv", "json", "txt"], "`train_file` should be a csv, a json or a txt file." + if self.validation_file is not None: + extension = self.validation_file.split(".")[-1] + assert extension in ["csv", "json", "txt"], "`validation_file` should be a csv, a json or a txt file." + + +def main(): + # See all possible arguments in src/transformers/training_args.py + # or by passing the --help flag to this script. + # We now keep distinct sets of args, for a cleaner separation of concerns. + + parser = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments)) + if len(sys.argv) == 2 and sys.argv[1].endswith(".json"): + # If we pass only one argument to the script and it's the path to a json file, + # let's parse it to get our arguments. + model_args, data_args, training_args = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1])) + else: + model_args, data_args, training_args = parser.parse_args_into_dataclasses() + + # Sending telemetry. Tracking the example usage helps us better allocate resources to maintain them. The + # information sent is the one passed as arguments along with your Python/PyTorch versions. + send_example_telemetry("run_clm", model_args, data_args) + + # Setup logging + logging.basicConfig( + format="%(asctime)s - %(levelname)s - %(name)s - %(message)s", + datefmt="%m/%d/%Y %H:%M:%S", + handlers=[logging.StreamHandler(sys.stdout)], + ) + + if training_args.should_log: + # The default of training_args.log_level is passive, so we set log level at info here to have that default. + transformers.utils.logging.set_verbosity_info() + + log_level = training_args.get_process_log_level() + logger.setLevel(log_level) + datasets.utils.logging.set_verbosity(log_level) + transformers.utils.logging.set_verbosity(log_level) + transformers.utils.logging.enable_default_handler() + transformers.utils.logging.enable_explicit_format() + + # Log on each process the small summary: + logger.warning( + f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}" + + f"distributed training: {bool(training_args.local_rank != -1)}, 16-bits training: {training_args.fp16}" + ) + logger.info(f"Training/evaluation parameters {training_args}") + + # Detecting last checkpoint. + last_checkpoint = None + if os.path.isdir(training_args.output_dir) and training_args.do_train and not training_args.overwrite_output_dir: + last_checkpoint = get_last_checkpoint(training_args.output_dir) + if last_checkpoint is None and len(os.listdir(training_args.output_dir)) > 0: + raise ValueError( + f"Output directory ({training_args.output_dir}) already exists and is not empty. " + "Use --overwrite_output_dir to overcome." + ) + elif last_checkpoint is not None and training_args.resume_from_checkpoint is None: + logger.info( + f"Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change " + "the `--output_dir` or add `--overwrite_output_dir` to train from scratch." + ) + + # Set seed before initializing model. + set_seed(training_args.seed) + + # Get the datasets: you can either provide your own CSV/JSON/TXT training and evaluation files (see below) + # or just provide the name of one of the public datasets available on the hub at https://huggingface.co/datasets/ + # (the dataset will be downloaded automatically from the datasets Hub). + # + # For CSV/JSON files, this script will use the column called 'text' or the first column if no column called + # 'text' is found. You can easily tweak this behavior (see below). + # + # In distributed training, the load_dataset function guarantee that only one local process can concurrently + # download the dataset. + if data_args.dataset_name is not None: + # Downloading and loading a dataset from the hub. + raw_datasets = load_dataset( + data_args.dataset_name, + data_args.dataset_config_name, + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + streaming=data_args.streaming, + ) + if "validation" not in raw_datasets.keys(): + raw_datasets["validation"] = load_dataset( + data_args.dataset_name, + data_args.dataset_config_name, + split=f"train[:{data_args.validation_split_percentage}%]", + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + streaming=data_args.streaming, + ) + raw_datasets["train"] = load_dataset( + data_args.dataset_name, + data_args.dataset_config_name, + split=f"train[{data_args.validation_split_percentage}%:]", + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + streaming=data_args.streaming, + ) + else: + data_files = {} + dataset_args = {} + if data_args.train_file is not None: + data_files["train"] = data_args.train_file + if data_args.validation_file is not None: + data_files["validation"] = data_args.validation_file + extension = ( + data_args.train_file.split(".")[-1] + if data_args.train_file is not None + else data_args.validation_file.split(".")[-1] + ) + if extension == "txt": + extension = "text" + dataset_args["keep_linebreaks"] = data_args.keep_linebreaks + raw_datasets = load_dataset( + extension, + data_files=data_files, + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + **dataset_args, + ) + # If no validation data is there, validation_split_percentage will be used to divide the dataset. + if "validation" not in raw_datasets.keys(): + raw_datasets["validation"] = load_dataset( + extension, + data_files=data_files, + split=f"train[:{data_args.validation_split_percentage}%]", + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + **dataset_args, + ) + raw_datasets["train"] = load_dataset( + extension, + data_files=data_files, + split=f"train[{data_args.validation_split_percentage}%:]", + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + **dataset_args, + ) + + # See more about loading any type of standard or custom dataset (from files, python dict, pandas DataFrame, etc) at + # https://huggingface.co/docs/datasets/loading_datasets.html. + + # Load pretrained model and tokenizer + # + # Distributed training: + # The .from_pretrained methods guarantee that only one local process can concurrently + # download model & vocab. + + config_kwargs = { + "cache_dir": model_args.cache_dir, + "revision": model_args.model_revision, + "use_auth_token": True if model_args.use_auth_token else None, + } + if model_args.config_name: + config = AutoConfig.from_pretrained(model_args.config_name, **config_kwargs) + elif model_args.model_name_or_path: + config = AutoConfig.from_pretrained(model_args.model_name_or_path, **config_kwargs) + else: + config = CONFIG_MAPPING[model_args.model_type]() + logger.warning("You are instantiating a new config instance from scratch.") + if model_args.config_overrides is not None: + logger.info(f"Overriding config: {model_args.config_overrides}") + config.update_from_string(model_args.config_overrides) + logger.info(f"New config: {config}") + + tokenizer_kwargs = { + "cache_dir": model_args.cache_dir, + "use_fast": model_args.use_fast_tokenizer, + "revision": model_args.model_revision, + "use_auth_token": True if model_args.use_auth_token else None, + } + if model_args.tokenizer_name: + tokenizer = LLaMATokenizer.from_pretrained(model_args.tokenizer_name, **tokenizer_kwargs) + elif model_args.model_name_or_path: + tokenizer = LLaMATokenizer.from_pretrained(model_args.model_name_or_path, **tokenizer_kwargs) + else: + raise ValueError( + "You are instantiating a new tokenizer from scratch. This is not supported by this script." + "You can do it from another script, save it, and load it from here, using --tokenizer_name." + ) + + if model_args.model_name_or_path: + torch_dtype = ( + model_args.torch_dtype + if model_args.torch_dtype in ["auto", None] + else getattr(torch, model_args.torch_dtype) + ) + model = LLaMAForCausalLM.from_pretrained( + model_args.model_name_or_path, + from_tf=bool(".ckpt" in model_args.model_name_or_path), + config=config, + cache_dir=model_args.cache_dir, + revision=model_args.model_revision, + use_auth_token=True if model_args.use_auth_token else None, + torch_dtype=torch_dtype, + ) + else: + model = LLaMAForCausalLM.from_config(config) + n_params = sum({p.data_ptr(): p.numel() for p in model.parameters()}.values()) + logger.info(f"Training new model from scratch - Total size={n_params/2**20:.2f}M params") + + # We resize the embeddings only when necessary to avoid index errors. If you are creating a model from scratch + # on a small vocab and want a smaller embedding size, remove this test. + embedding_size = model.get_input_embeddings().weight.shape[0] + if len(tokenizer) > embedding_size: + model.resize_token_embeddings(len(tokenizer)) + + # Preprocessing the datasets. + # First we tokenize all the texts. + if training_args.do_train: + column_names = list(raw_datasets["train"].features) + else: + column_names = list(raw_datasets["validation"].features) + text_column_name = "text" if "text" in column_names else column_names[0] + + # since this will be pickled to avoid _LazyModule error in Hasher force logger loading before tokenize_function + tok_logger = transformers.utils.logging.get_logger("transformers.tokenization_utils_base") + + def tokenize_function(examples): + with CaptureLogger(tok_logger) as cl: + output = tokenizer(examples[text_column_name]) + # clm input could be much much longer than block_size + if "Token indices sequence length is longer than the" in cl.out: + tok_logger.warning( + "^^^^^^^^^^^^^^^^ Please ignore the warning above - this long input will be chunked into smaller bits" + " before being passed to the model." + ) + return output + + with training_args.main_process_first(desc="dataset map tokenization"): + if not data_args.streaming: + tokenized_datasets = raw_datasets.map( + tokenize_function, + batched=True, + num_proc=data_args.preprocessing_num_workers, + remove_columns=column_names, + load_from_cache_file=not data_args.overwrite_cache, + desc="Running tokenizer on dataset", + ) + else: + tokenized_datasets = raw_datasets.map( + tokenize_function, + batched=True, + remove_columns=column_names, + ) + + if data_args.block_size is None: + block_size = tokenizer.model_max_length + if block_size > 1024: + logger.warning( + "The chosen tokenizer supports a `model_max_length` that is longer than the default `block_size` value" + " of 1024. If you would like to use a longer `block_size` up to `tokenizer.model_max_length` you can" + " override this default with `--block_size xxx`." + ) + block_size = 1024 + else: + if data_args.block_size > tokenizer.model_max_length: + logger.warning( + f"The block_size passed ({data_args.block_size}) is larger than the maximum length for the model" + f"({tokenizer.model_max_length}). Using block_size={tokenizer.model_max_length}." + ) + block_size = min(data_args.block_size, tokenizer.model_max_length) + + # Main data processing function that will concatenate all texts from our dataset and generate chunks of block_size. + def group_texts(examples): + # Concatenate all texts. + if examples["input"]: + return f"""Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request. + +### Instruction: +{examples["instruction"]} + +### Input: +{examples["input"]} + +### Response: +{examples["output"]}""" + else: + return f"""Below is an instruction that describes a task. Write a response that appropriately completes the request. + +### Instruction: +{examples["instruction"]} + +### Response: +{examples["output"]}""" + + concatenated_examples = {k: list(chain(*examples[k])) for k in examples.keys()} + total_length = len(concatenated_examples[list(examples.keys())[0]]) + # We drop the small remainder, we could add padding if the model supported it instead of this drop, you can + # customize this part to your needs. + if total_length >= block_size: + total_length = (total_length // block_size) * block_size + # Split by chunks of max_len. + result = { + k: [t[i : i + block_size] for i in range(0, total_length, block_size)] + for k, t in concatenated_examples.items() + } + result["labels"] = result["input_ids"].copy() + return result + + # Note that with `batched=True`, this map processes 1,000 texts together, so group_texts throws away a remainder + # for each of those groups of 1,000 texts. You can adjust that batch_size here but a higher value might be slower + # to preprocess. + # + # To speed up this part, we use multiprocessing. See the documentation of the map method for more information: + # https://huggingface.co/docs/datasets/package_reference/main_classes.html#datasets.Dataset.map + + with training_args.main_process_first(desc="grouping texts together"): + if not data_args.streaming: + lm_datasets = tokenized_datasets.map( + group_texts, + batched=True, + num_proc=data_args.preprocessing_num_workers, + load_from_cache_file=not data_args.overwrite_cache, + desc=f"Grouping texts in chunks of {block_size}", + ) + else: + lm_datasets = tokenized_datasets.map( + group_texts, + batched=True, + ) + + if training_args.do_train: + if "train" not in tokenized_datasets: + raise ValueError("--do_train requires a train dataset") + train_dataset = lm_datasets["train"] + if data_args.max_train_samples is not None: + max_train_samples = min(len(train_dataset), data_args.max_train_samples) + train_dataset = train_dataset.select(range(max_train_samples)) + + if training_args.do_eval: + if "validation" not in tokenized_datasets: + raise ValueError("--do_eval requires a validation dataset") + eval_dataset = lm_datasets["validation"] + if data_args.max_eval_samples is not None: + max_eval_samples = min(len(eval_dataset), data_args.max_eval_samples) + eval_dataset = eval_dataset.select(range(max_eval_samples)) + + def preprocess_logits_for_metrics(logits, labels): + if isinstance(logits, tuple): + # Depending on the model and config, logits may contain extra tensors, + # like past_key_values, but logits always come first + logits = logits[0] + return logits.argmax(dim=-1) + + metric = evaluate.load("accuracy") + + def compute_metrics(eval_preds): + preds, labels = eval_preds + # preds have the same shape as the labels, after the argmax(-1) has been calculated + # by preprocess_logits_for_metrics but we need to shift the labels + labels = labels[:, 1:].reshape(-1) + preds = preds[:, :-1].reshape(-1) + return metric.compute(predictions=preds, references=labels) + + # Initialize our Trainer + trainer = Trainer( + model=model, + args=training_args, + train_dataset=train_dataset if training_args.do_train else None, + eval_dataset=eval_dataset if training_args.do_eval else None, + tokenizer=tokenizer, + # Data collator will default to DataCollatorWithPadding, so we change it. + data_collator=default_data_collator, + compute_metrics=compute_metrics if training_args.do_eval and not is_torch_tpu_available() else None, + preprocess_logits_for_metrics=preprocess_logits_for_metrics + if training_args.do_eval and not is_torch_tpu_available() + else None, + ) + + # Training + if training_args.do_train: + checkpoint = None + if training_args.resume_from_checkpoint is not None: + checkpoint = training_args.resume_from_checkpoint + elif last_checkpoint is not None: + checkpoint = last_checkpoint + train_result = trainer.train(resume_from_checkpoint=checkpoint) + trainer.save_model() # Saves the tokenizer too for easy upload + + metrics = train_result.metrics + + max_train_samples = ( + data_args.max_train_samples if data_args.max_train_samples is not None else len(train_dataset) + ) + metrics["train_samples"] = min(max_train_samples, len(train_dataset)) + + trainer.log_metrics("train", metrics) + trainer.save_metrics("train", metrics) + trainer.save_state() + + # Evaluation + if training_args.do_eval: + logger.info("*** Evaluate ***") + + metrics = trainer.evaluate() + + max_eval_samples = data_args.max_eval_samples if data_args.max_eval_samples is not None else len(eval_dataset) + metrics["eval_samples"] = min(max_eval_samples, len(eval_dataset)) + try: + perplexity = math.exp(metrics["eval_loss"]) + except OverflowError: + perplexity = float("inf") + metrics["perplexity"] = perplexity + + trainer.log_metrics("eval", metrics) + trainer.save_metrics("eval", metrics) + + kwargs = {"finetuned_from": model_args.model_name_or_path, "tasks": "text-generation"} + if data_args.dataset_name is not None: + kwargs["dataset_tags"] = data_args.dataset_name + if data_args.dataset_config_name is not None: + kwargs["dataset_args"] = data_args.dataset_config_name + kwargs["dataset"] = f"{data_args.dataset_name} {data_args.dataset_config_name}" + else: + kwargs["dataset"] = data_args.dataset_name + + if training_args.push_to_hub: + trainer.push_to_hub(**kwargs) + else: + trainer.create_model_card(**kwargs) + + +def _mp_fn(index): + # For xla_spawn (TPUs) + main() + + +if __name__ == "__main__": + main() \ No newline at end of file diff --git a/train.sh b/train.sh new file mode 100644 index 0000000..9ddab58 --- /dev/null +++ b/train.sh @@ -0,0 +1,8 @@ +python train.py \ + --model_name_or_path decapoda-research/llama-7b-hf \ + --dataset_name alpaca_data.json \ + --per_device_train_batch_size 8 \ + --per_device_eval_batch_size 8 \ + --do_train \ + --do_eval \ + --output_dir test-clm \ No newline at end of file