gpt4all/train.py

206 lines
7.2 KiB
Python
Raw Normal View History

2023-03-25 12:17:48 -04:00
import os
from transformers import AutoModelForCausalLM, AutoTokenizer
from transformers.trainer_pt_utils import get_parameter_names
import torch
import torch.nn as nn
from argparse import ArgumentParser
from read import read_config
from accelerate import Accelerator
from accelerate.utils import DummyScheduler, DummyOptim, set_seed
from peft import get_peft_model, LoraConfig, TaskType
from data import load_data
from torchmetrics import MeanMetric
from tqdm import tqdm
def format_metrics(metrics, split, prefix=""):
log = f"[{split}]" + prefix
log += " ".join([f"{key}: {value:.4f}" for key, value in metrics.items()])
return log
def evaluate(config, model, val_dataloader):
model.eval()
val_loss = MeanMetric().to(model.device)
with torch.no_grad():
for i, batch in enumerate(
tqdm(val_dataloader),
):
if i == config["eval_steps"]:
break
loss = model(**batch).loss
loss_values = accelerator.gather_for_metrics({"loss": loss.detach()})
val_loss.update(loss_values["loss"])
return val_loss
def train(accelerator, config):
set_seed(config['seed'])
accelerator.print(config)
accelerator.print(f"Using {accelerator.num_processes} GPUs")
tokenizer = AutoTokenizer.from_pretrained(config['tokenizer_name'])
2023-03-26 13:45:21 -04:00
# llama has no pad token, set it to new token
2023-03-25 12:17:48 -04:00
if tokenizer.pad_token is None:
# these tokens are already in the vocab, just not mapped correctly
2023-03-26 13:45:21 -04:00
added_tokens = tokenizer.add_special_tokens({"bos_token": "<s>", "eos_token": "</s>", "pad_token": "<pad>"})
2023-03-25 12:17:48 -04:00
with accelerator.main_process_first():
train_dataloader, val_dataloader = load_data(config, tokenizer)
2023-03-27 13:30:24 -04:00
2023-03-25 12:17:48 -04:00
checkpoint = config["gradient_checkpointing"]
model = AutoModelForCausalLM.from_pretrained(config["model_name"],
use_cache=False if checkpoint else True,
trust_remote_code=True)
2023-03-26 13:45:21 -04:00
if added_tokens > 0:
model.resize_token_embeddings(len(tokenizer))
2023-03-25 12:17:48 -04:00
if checkpoint:
model.gradient_checkpointing_enable()
if config["lora"]:
peft_config = LoraConfig(
# should R be configurable?
task_type=TaskType.CAUSAL_LM, inference_mode=False, r=8, lora_alpha=32, lora_dropout=0.1
)
model = get_peft_model(model, peft_config)
model.print_trainable_parameters()
optimizer_cls = (
torch.optim.AdamW
if accelerator.state.deepspeed_plugin is None
or "optimizer" not in accelerator.state.deepspeed_plugin.deepspeed_config
else DummyOptim
)
# karpathy doesn't decay embeddding, maybe we should exclude
# https://github.com/karpathy/minGPT/commit/bbbdac74fa9b2e55574d70056163ffbae42310c1#diff-2075fa9c224b395be5bda85544dd36572b59c76c54562819eadadbf268602834R157s
optimizer = optimizer_cls(model.parameters(), lr=config["lr"])
# scheduler defined in Deepspeed config
scheduler = DummyScheduler(
optimizer, warmup_num_steps=config["warmup_steps"],
)
model, optimizer, train_dataloader, val_dataloader, scheduler = accelerator.prepare(
model, optimizer, train_dataloader, val_dataloader, scheduler
)
# setup for saving training states in case preemption
accelerator.register_for_checkpointing(scheduler)
if config["checkpoint"]:
accelerator.load_state(config["checkpoint"])
accelerator.print(f"Resumed from checkpoint: {config['checkpoint']}")
path = os.path.basename(config["train_args"]["resume_from_checkpoint"])
training_difference = os.path.splitext(path)[0]
resume_step = int(training_difference.replace("step_", ""))
accelerator.skip_first_batches(train_dataloader, resume_step)
accelerator.print(f"Resuming from step {resume_step}")
train_loss = MeanMetric().to(model.device)
2023-03-26 13:45:21 -04:00
if accelerator.state.deepspeed_plugin is not None:
gradient_accumulation_steps = accelerator.state.deepspeed_plugin.deepspeed_config[
"gradient_accumulation_steps"
]
2023-03-27 12:32:35 -04:00
for epoch in range(config["num_epochs"]):
for step, batch in enumerate(tqdm(train_dataloader)):
model.train()
outputs = model(**batch)
loss = outputs.loss
loss = loss / gradient_accumulation_steps
2023-03-25 12:17:48 -04:00
2023-03-27 12:32:35 -04:00
accelerator.backward(loss)
2023-03-25 12:17:48 -04:00
2023-03-27 12:32:35 -04:00
# log LR in case something weird happens
if step > 0 and step % (config["eval_every"] // 10) == 0:
if config["wandb"]:
accelerator.log({"lr": scheduler.get_last_lr()[0]}, step=step)
2023-03-25 12:17:48 -04:00
2023-03-27 12:32:35 -04:00
if (step + 1) % gradient_accumulation_steps == 0 or step == len(train_dataloader) - 1:
optimizer.step()
scheduler.step()
optimizer.zero_grad()
2023-03-25 12:17:48 -04:00
2023-03-27 12:32:35 -04:00
loss_values = accelerator.gather_for_metrics({"loss": loss.detach()})
train_loss.update(loss_values["loss"])
2023-03-25 12:17:48 -04:00
2023-03-27 12:32:35 -04:00
if step > 0 and step % config["save_every"] == 0:
accelerator.save_state(f"{config['output_dir']}/step_{step}")
2023-03-25 12:17:48 -04:00
2023-03-27 12:32:35 -04:00
if step > 0 and step % config["eval_every"] == 0:
val_loss = evaluate(config, model, val_dataloader)
2023-03-25 12:17:48 -04:00
2023-03-27 12:32:35 -04:00
log_train = {
"train_loss": train_loss.compute()
}
log_val = {
"val_loss": val_loss.compute()
2023-03-25 12:17:48 -04:00
}
2023-03-27 12:32:35 -04:00
if config["wandb"]:
accelerator.log({**log_train, **log_val}, step=step)
accelerator.print(f"Current LR: {scheduler.get_last_lr()[0]}")
accelerator.print(format_metrics(log_train, "train", f" step {step} "))
accelerator.print(format_metrics(log_val, "val", f" step {step} "))
2023-03-25 12:17:48 -04:00
2023-03-27 12:32:35 -04:00
train_loss.reset()
2023-03-25 12:17:48 -04:00
2023-03-27 12:32:35 -04:00
accelerator.print(f"Epoch {epoch} finished")
accelerator.print(f"Pushing to HF hub")
accelerator.wait_for_everyone()
unwrapped_model = accelerator.unwrap_model(model)
if accelerator.is_main_process:
2023-03-27 13:30:24 -04:00
unwrapped_model.push_to_hub(config["save_name"] + "_first_epoch", private=True)
2023-03-25 12:17:48 -04:00
accelerator.wait_for_everyone()
unwrapped_model = accelerator.unwrap_model(model)
unwrapped_model.save_pretrained(
f"{config['output_dir']}/final",
is_main_process=accelerator.is_main_process,
save_function=accelerator.save,
state_dict=accelerator.get_state_dict(model),
)
2023-03-27 12:32:35 -04:00
if accelerator.is_main_process:
unwrapped_model.push_to_hub(config["save_name"], private=True)
2023-03-25 12:17:48 -04:00
accelerator.end_training()
if __name__ == "__main__":
# parse arguments by reading in a config
parser = ArgumentParser()
parser.add_argument("--config", type=str, default="config.yaml")
args = parser.parse_args()
config = read_config(args.config)
if config["wandb"]:
accelerator = Accelerator(log_with="wandb")
accelerator.init_trackers(
project_name=config["wandb_project_name"],
config=config,
init_kwargs={"wandb": {"entity": config["wandb_entity"]}},
)
else:
accelerator = Accelerator()
train(accelerator, config=config)