import os from transformers import AutoModelForCausalLM, AutoTokenizer, get_scheduler import torch from torch.optim import AdamW 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 import wandb torch.backends.cuda.matmul.allow_tf32 = True 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(model, val_dataloader): model.eval() val_loss = MeanMetric(nan_strategy="error").to(model.device) with torch.no_grad(): for batch in tqdm(val_dataloader): 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'], model_max_length=config['max_length'], use_fast=False) # if no pad token, set it to eos if tokenizer.pad_token is None: tokenizer.pad_token = tokenizer.eos_token with accelerator.main_process_first(): train_dataloader, val_dataloader = load_data(config, tokenizer) checkpoint = config["gradient_checkpointing"] model = AutoModelForCausalLM.from_pretrained(config["model_name"], use_cache=False if checkpoint else True, trust_remote_code=True) 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 = ( 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"], weight_decay=config["weight_decay"]) if accelerator.state.deepspeed_plugin is not None: gradient_accumulation_steps = accelerator.state.deepspeed_plugin.deepspeed_config[ "gradient_accumulation_steps" ] # decay to min_lr instead of 0 lr_ratio = config["min_lr"] / config["lr"] accelerator.print(f"Len of train_dataloader: {len(train_dataloader)}") total_num_steps = (len(train_dataloader) / gradient_accumulation_steps) * (config["num_epochs"]) # instead of decaying to zero, decay to ratio of min_lr / lr total_num_steps += int(total_num_steps * lr_ratio) + config["warmup_steps"] accelerator.print(f"Total training steps: {total_num_steps}") # Creates Dummy Scheduler if `scheduler` was specified in the config file else creates `args.lr_scheduler_type` Scheduler if ( accelerator.state.deepspeed_plugin is None or "scheduler" not in accelerator.state.deepspeed_plugin.deepspeed_config ): scheduler = get_scheduler( name="cosine", optimizer=optimizer, num_warmup_steps=config["warmup_steps"] * accelerator.num_processes, num_training_steps=total_num_steps, ) else: scheduler = DummyScheduler( optimizer, total_num_steps=total_num_steps, 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["checkpoint"]) training_difference = os.path.splitext(path)[0] resume_step = int(training_difference.replace("step_", "")) train_dataloader = accelerator.skip_first_batches(train_dataloader, resume_step) accelerator.print(f"Resuming from step {resume_step}") else: resume_step = 0 # log gradients if accelerator.is_main_process and config["wandb"]: wandb.watch(model, log_freq=config["log_grads_every"], log="all") accelerator.wait_for_everyone() for epoch in range(0, config["num_epochs"]): train_loss = MeanMetric(nan_strategy="error").to(model.device) for step, batch in enumerate(tqdm(train_dataloader)): curr_step = epoch * len(train_dataloader) + step model.train() outputs = model(**batch) loss = outputs.loss # gather loss before backprop in case of gradient accumulation loss_values = accelerator.gather_for_metrics({"loss": loss.detach().float()}) if config["wandb"]: accelerator.log({"loss": torch.mean(loss_values["loss"]).item()}, step=curr_step) train_loss.update(loss_values["loss"]) loss = loss / gradient_accumulation_steps accelerator.backward(loss) # get gradient norm of all params # log LR in case something weird happens if step > 0 and step % (config["log_lr_every"]) == 0: if config["wandb"]: accelerator.log({"lr": scheduler.get_last_lr()[0]}, step=curr_step) if (step + 1) % gradient_accumulation_steps == 0 or step == len(train_dataloader) - 1: optimizer.step() scheduler.step() optimizer.zero_grad() if step > 0 and step % config["save_every"] == 0: accelerator.save_state(f"{config['output_dir']}/step_{curr_step}") if step > 0 and (step % config["eval_every"] == 0 or step == len(train_dataloader) - 1): val_loss = evaluate(model, val_dataloader) log_train = { "train_loss": train_loss.compute() } log_val = { "val_loss": val_loss.compute() } if config["wandb"]: accelerator.log({**log_train, **log_val}, step=curr_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} ")) train_loss.reset() accelerator.print(f"Epoch {epoch} finished") accelerator.print(f"Pushing to HF hub") unwrapped_model = accelerator.unwrap_model(model) unwrapped_model.save_pretrained( f"{config['output_dir']}/epoch_{epoch}", is_main_process=accelerator.is_main_process, save_function=accelerator.save, state_dict=accelerator.get_state_dict(model), ) try: if accelerator.is_main_process: unwrapped_model.push_to_hub(config["save_name"] + f"-epoch_{epoch}", private=True) except Exception as e: accelerator.print(e) accelerator.print(f"Failed to push to hub") if config["num_epochs"] > 1: 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), ) 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)