reconstruct chat trainer and fix training script (#3588)

Co-authored-by: Yuanchen Xu <yuanchen.xu00@gmail.com>

applications/Chat/benchmarks/benchmark_gpt_dummy.py

@@ -156,8 +156,10 @@ def main(args):
                          eos_token_id=tokenizer.eos_token_id,
                          callbacks=[performance_evaluator])
 
-    random_prompts = torch.randint(tokenizer.vocab_size, (1000, 400), device=torch.cuda.current_device())
+    random_prompts = torch.randint(tokenizer.vocab_size, (1000, 1, 400), device=torch.cuda.current_device())
-    trainer.fit(random_prompts,
+    random_attention_mask = torch.randint(1, (1000, 1, 400), device=torch.cuda.current_device()).to(torch.bool)
+    random_pretrain = [{'input_ids':random_prompts[i], 'labels':random_prompts[i], 'attention_mask':random_attention_mask[i]} for i in range(1000)]
+    trainer.fit(random_prompts, random_pretrain,
                 num_episodes=args.num_episodes,
                 max_timesteps=args.max_timesteps,
                 update_timesteps=args.update_timesteps)

applications/Chat/benchmarks/benchmark_opt_lora_dummy.py

@@ -149,8 +149,10 @@ def main(args):
                          eos_token_id=tokenizer.eos_token_id,
                          callbacks=[performance_evaluator])
 
-    random_prompts = torch.randint(tokenizer.vocab_size, (1000, 400), device=torch.cuda.current_device())
+    random_prompts = torch.randint(tokenizer.vocab_size, (1000, 1, 400), device=torch.cuda.current_device())
-    trainer.fit(random_prompts,
+    random_attention_mask = torch.randint(1, (1000, 1, 400), device=torch.cuda.current_device()).to(torch.bool)
+    random_pretrain = [{'input_ids':random_prompts[i], 'labels':random_prompts[i], 'attention_mask':random_attention_mask[i]} for i in range(1000)]
+    trainer.fit(random_prompts, random_pretrain,
                 num_episodes=args.num_episodes,
                 max_timesteps=args.max_timesteps,
                 update_timesteps=args.update_timesteps)

applications/Chat/coati/trainer/base.py

@@ -2,15 +2,10 @@ from abc import ABC, abstractmethod
 from typing import Any, Callable, Dict, List, Optional, Union
 
 import torch
-from coati.experience_maker import Experience, ExperienceMaker
+from coati.experience_maker import Experience
-from coati.replay_buffer import ReplayBuffer
-from torch import Tensor
-from torch.utils.data import DistributedSampler
-from tqdm import tqdm
 
 from .callbacks import Callback
 from .strategies import Strategy
-from .utils import is_rank_0
 
 
 class Trainer(ABC):
@@ -19,113 +14,28 @@ class Trainer(ABC):
 
     Args:
         strategy (Strategy):the strategy to use for training
-        experience_maker (ExperienceMaker): the experience maker to use for produce experience to fullfill replay buffer
-        replay_buffer (ReplayBuffer): the replay buffer to use for training
-        experience_batch_size (int, defaults to 8): the batch size to use for experience generation
         max_epochs (int, defaults to 1): the number of epochs of training process
         tokenizer (Callable, optional): the tokenizer to use for tokenizing the input
-        sample_replay_buffer (bool, defaults to False): whether to sample from replay buffer
+        dataloader_pin_memory (bool, defaults to True): whether to pin memory for data loader
-        data_loader_pin_memory (bool, defaults to True): whether to pin memory for data loader
         callbacks (List[Callback], defaults to []): the callbacks to call during training process
         generate_kwargs (dict, optional): the kwargs to use while model generating
     """
 
     def __init__(self,
                  strategy: Strategy,
-                 experience_maker: ExperienceMaker,
-                 replay_buffer: ReplayBuffer,
-                 experience_batch_size: int = 8,
                  max_epochs: int = 1,
                  tokenizer: Optional[Callable[[Any], dict]] = None,
-                 sample_replay_buffer: bool = False,
                  dataloader_pin_memory: bool = True,
                  callbacks: List[Callback] = [],
                  **generate_kwargs) -> None:
         super().__init__()
         self.strategy = strategy
-        self.experience_maker = experience_maker
-        self.replay_buffer = replay_buffer
-        self.experience_batch_size = experience_batch_size
         self.max_epochs = max_epochs
         self.tokenizer = tokenizer
         self.generate_kwargs = generate_kwargs
-        self.sample_replay_buffer = sample_replay_buffer
         self.dataloader_pin_memory = dataloader_pin_memory
         self.callbacks = callbacks
 
-    @abstractmethod
-    def training_step(self, experience: Experience) -> Dict[str, Any]:
-        pass
-
-    def _make_experience(self, inputs: Union[Tensor, Dict[str, Tensor]]) -> Experience:
-        if isinstance(inputs, Tensor):
-            return self.experience_maker.make_experience(inputs, **self.generate_kwargs)
-        elif isinstance(inputs, dict):
-            return self.experience_maker.make_experience(**inputs, **self.generate_kwargs)
-        else:
-            raise ValueError(f'Unsupported input type "{type(inputs)}"')
-
-    def _sample_prompts(self, prompts) -> list:
-        indices = list(range(len(prompts)))
-        sampled_indices = self.strategy.experience_sampler.choice(indices, self.experience_batch_size, replace=False)
-        return [prompts[i] for i in sampled_indices]
-
-    def _learn(self):
-        # replay buffer may be empty at first, we should rebuild at each training
-        if not self.sample_replay_buffer:
-            dataloader = self.strategy.setup_dataloader(self.replay_buffer, self.dataloader_pin_memory)
-            device = torch.cuda.current_device()
-        if self.sample_replay_buffer:
-            pbar = tqdm(range(self.max_epochs), desc='Train epoch', disable=not is_rank_0())
-            for _ in pbar:
-                experience = self.replay_buffer.sample()
-                metrics = self.training_step(experience)
-                pbar.set_postfix(metrics)
-        else:
-            for epoch in range(self.max_epochs):
-                self._on_learn_epoch_start(epoch)
-                if isinstance(dataloader.sampler, DistributedSampler):
-                    dataloader.sampler.set_epoch(epoch)
-                pbar = tqdm(dataloader, desc=f'Train epoch [{epoch+1}/{self.max_epochs}]', disable=not is_rank_0())
-                for experience in pbar:
-                    self._on_learn_batch_start()
-                    experience.to_device(device)
-                    metrics = self.training_step(experience)
-                    self._on_learn_batch_end(metrics, experience)
-                    pbar.set_postfix(metrics)
-                self._on_learn_epoch_end(epoch)
-
-    def fit(self,
-            prompt_dataloader,
-            pretrain_dataloader,
-            num_episodes: int = 50000,
-            max_timesteps: int = 500,
-            update_timesteps: int = 5000) -> None:
-        time = 0
-        self.pretrain_dataloader = pretrain_dataloader
-        self.prompt_dataloader = prompt_dataloader
-        self._on_fit_start()
-        for episode in range(num_episodes):
-            self._on_episode_start(episode)
-            for timestep in tqdm(range(max_timesteps),
-                                 desc=f'Episode [{episode+1}/{num_episodes}]',
-                                 disable=not is_rank_0()):
-                time += 1
-                prompts = next(iter(self.prompt_dataloader))
-                self._on_make_experience_start()
-                self.experience_maker.initial_model.to(torch.cuda.current_device())
-                self.experience_maker.reward_model.to(torch.cuda.current_device())
-                experience = self._make_experience(prompts)
-                self._on_make_experience_end(experience)
-                self.replay_buffer.append(experience)
-                if time % update_timesteps == 0:
-                    self.experience_maker.initial_model.to('cpu')
-                    self.experience_maker.reward_model.to('cpu')
-                    self._learn()
-                    self.replay_buffer.clear()
-            self._on_episode_end(episode)
-        self._on_fit_end()
-
     # TODO(ver217): maybe simplify these code using context
     def _on_fit_start(self) -> None:
         for callback in self.callbacks:

applications/Chat/coati/trainer/ppo.py

@@ -1,4 +1,4 @@
-from typing import Any, Callable, Dict, List, Optional
+from typing import Any, Callable, Dict, List, Optional, Union
 
 import torch
 import torch.nn as nn
@@ -7,12 +7,16 @@ from coati.models.base import Actor, Critic
 from coati.models.generation_utils import update_model_kwargs_fn
 from coati.models.loss import PolicyLoss, ValueLoss
 from coati.replay_buffer import NaiveReplayBuffer
+from torch import Tensor
 from torch.optim import Optimizer
+from torch.utils.data import DistributedSampler
 from transformers.tokenization_utils_base import PreTrainedTokenizerBase
+from tqdm import tqdm
 
 from .base import Trainer
 from .callbacks import Callback
 from .strategies import Strategy
+from .utils import is_rank_0
 
 
 class PPOTrainer(Trainer):
@@ -33,6 +37,7 @@ class PPOTrainer(Trainer):
         buffer_cpu_offload (bool, defaults to True): whether to offload replay buffer to cpu
         eps_clip (float, defaults to 0.2): the clip coefficient of policy loss
         vf_coef (float, defaults to 1.0): the coefficient of value loss
+        ptx_coef (float, defaults to 0.9): the coefficient of ptx loss
         value_clip (float, defaults to 0.4): the clip coefficient of value loss
         experience_batch_size (int, defaults to 8): the batch size to use for experience generation
         max_epochs (int, defaults to 1): the number of epochs of training process
@@ -69,8 +74,13 @@ class PPOTrainer(Trainer):
         experience_maker = NaiveExperienceMaker(actor, critic, reward_model, initial_model, kl_coef)
         replay_buffer = NaiveReplayBuffer(train_batch_size, buffer_limit, buffer_cpu_offload)
         generate_kwargs = _set_default_generate_kwargs(strategy, generate_kwargs, actor)
-        super().__init__(strategy, experience_maker, replay_buffer, experience_batch_size, max_epochs, tokenizer,
+        super().__init__(strategy, max_epochs, tokenizer, dataloader_pin_memory, callbacks, **generate_kwargs)
-                         sample_replay_buffer, dataloader_pin_memory, callbacks, **generate_kwargs)
+
+        self.experience_maker = experience_maker
+        self.replay_buffer = replay_buffer
+        self.experience_batch_size = experience_batch_size
+        self.sample_replay_buffer = sample_replay_buffer
+
         self.actor = actor
         self.critic = critic
 
@@ -82,6 +92,81 @@ class PPOTrainer(Trainer):
         self.actor_optim = actor_optim
         self.critic_optim = critic_optim
 
+    def _make_experience(self, inputs: Union[Tensor, Dict[str, Tensor]]) -> Experience:
+        if isinstance(inputs, Tensor):
+            return self.experience_maker.make_experience(inputs, **self.generate_kwargs)
+        elif isinstance(inputs, dict):
+            return self.experience_maker.make_experience(**inputs, **self.generate_kwargs)
+        else:
+            raise ValueError(f'Unsupported input type "{type(inputs)}"')
+
+    def _sample_prompts(self, prompts) -> list:
+        indices = list(range(len(prompts)))
+        sampled_indices = self.strategy.experience_sampler.choice(
+            indices, self.experience_batch_size, replace=False)
+        return [prompts[i] for i in sampled_indices]
+
+    def _learn(self):
+        # replay buffer may be empty at first, we should rebuild at each training
+        if not self.sample_replay_buffer:
+            dataloader = self.strategy.setup_dataloader(
+                self.replay_buffer, self.dataloader_pin_memory)
+            device = torch.cuda.current_device()
+        if self.sample_replay_buffer:
+            pbar = tqdm(range(self.max_epochs), desc='Train epoch',
+                        disable=not is_rank_0())
+            for _ in pbar:
+                experience = self.replay_buffer.sample()
+                metrics = self.training_step(experience)
+                pbar.set_postfix(metrics)
+        else:
+            for epoch in range(self.max_epochs):
+                self._on_learn_epoch_start(epoch)
+                if isinstance(dataloader.sampler, DistributedSampler):
+                    dataloader.sampler.set_epoch(epoch)
+                pbar = tqdm(
+                    dataloader, desc=f'Train epoch [{epoch+1}/{self.max_epochs}]', disable=not is_rank_0())
+                for experience in pbar:
+                    self._on_learn_batch_start()
+                    experience.to_device(device)
+                    metrics = self.training_step(experience)
+                    self._on_learn_batch_end(metrics, experience)
+                    pbar.set_postfix(metrics)
+                self._on_learn_epoch_end(epoch)
+
+    def fit(self,
+            prompt_dataloader,
+            pretrain_dataloader,
+            num_episodes: int = 50000,
+            max_timesteps: int = 500,
+            update_timesteps: int = 5000) -> None:
+        time = 0
+        self.pretrain_dataloader = pretrain_dataloader
+        self.prompt_dataloader = prompt_dataloader
+        self._on_fit_start()
+        for episode in range(num_episodes):
+            self._on_episode_start(episode)
+            for timestep in tqdm(range(max_timesteps),
+                                 desc=f'Episode [{episode+1}/{num_episodes}]',
+                                 disable=not is_rank_0()):
+                time += 1
+                prompts = next(iter(self.prompt_dataloader))
+                self._on_make_experience_start()
+                self.experience_maker.initial_model.to(
+                    torch.cuda.current_device())
+                self.experience_maker.reward_model.to(
+                    torch.cuda.current_device())
+                experience = self._make_experience(prompts)
+                self._on_make_experience_end(experience)
+                self.replay_buffer.append(experience)
+                if time % update_timesteps == 0:
+                    self.experience_maker.initial_model.to('cpu')
+                    self.experience_maker.reward_model.to('cpu')
+                    self._learn()
+                    self.replay_buffer.clear()
+            self._on_episode_end(episode)
+        self._on_fit_end()
+
     def training_step(self, experience: Experience) -> Dict[str, float]:
         self.actor.train()
         self.critic.train()

applications/Chat/coati/trainer/rm.py

@@ -1,6 +1,5 @@
-from abc import ABC
 from datetime import datetime
-from typing import Optional
+from typing import Optional, List
 
 import pandas as pd
 import torch
@@ -10,11 +9,13 @@ from torch.utils.data import DataLoader, Dataset, DistributedSampler
 from tqdm import tqdm
 from transformers.tokenization_utils_base import PreTrainedTokenizerBase
 
+from .callbacks import Callback
+from .base import Trainer
 from .strategies import Strategy
 from .utils import is_rank_0
 
 
-class RewardModelTrainer(ABC):
+class RewardModelTrainer(Trainer):
     """
         Trainer to use while training reward model.
 
@@ -23,11 +24,12 @@ class RewardModelTrainer(ABC):
         strategy (Strategy): the strategy to use for training
         optim(Optimizer): the optimizer to use for training
         loss_fn (callable): the loss function to use for training
-        train_dataset (Dataset): the dataset to use for training
+        train_dataloader (DataLoader): the dataloader to use for training
-        valid_dataset (Dataset): the dataset to use for validation
+        valid_dataloader (DataLoader): the dataloader to use for validation
-        eval_dataset (Dataset): the dataset to use for evaluation
+        eval_dataloader (DataLoader): the dataloader to use for evaluation
         batch_size (int, defaults to 1): the batch size while training
         max_epochs (int, defaults to 2): the number of epochs to train
+        callbacks (List[Callback], defaults to []): the callbacks to call during training process
     """
 
     def __init__(
@@ -36,25 +38,19 @@ class RewardModelTrainer(ABC):
         strategy: Strategy,
         optim: Optimizer,
         loss_fn,
-        train_dataset: Dataset,
+        train_dataloader: DataLoader,
-        valid_dataset: Dataset,
+        valid_dataloader: DataLoader,
-        eval_dataset: Dataset,
+        eval_dataloader: DataLoader,
         batch_size: int = 1,
         max_epochs: int = 1,
+        callbacks: List[Callback] = [],
     ) -> None:
-        super().__init__()
+        super().__init__(strategy, max_epochs, callbacks=callbacks)
-        self.strategy = strategy
-        self.epochs = max_epochs
         train_sampler = None
 
-        if dist.is_initialized() and dist.get_world_size() > 1:
+        self.train_dataloader = train_dataloader
-            train_sampler = DistributedSampler(train_dataset, shuffle=True, seed=42, drop_last=True)
+        self.valid_dataloader = valid_dataloader
-        self.train_dataloader = DataLoader(train_dataset,
+        self.eval_dataloader = eval_dataloader
-                                           shuffle=(train_sampler is None),
-                                           sampler=train_sampler,
-                                           batch_size=batch_size)
-        self.valid_dataloader = DataLoader(valid_dataset, batch_size=batch_size, shuffle=True)
-        self.eval_dataloader = DataLoader(eval_dataset, batch_size=batch_size, shuffle=True)
 
         self.model = strategy.setup_model(model)
         self.loss_fn = loss_fn
@@ -86,8 +82,8 @@ class RewardModelTrainer(ABC):
 
     def fit(self):
         time = datetime.now()
-        epoch_bar = tqdm(range(self.epochs), desc='Train epoch', disable=not is_rank_0())
+        epoch_bar = tqdm(range(self.max_epochs), desc='Train epoch', disable=not is_rank_0())
-        for epoch in range(self.epochs):
+        for epoch in range(self.max_epochs):
             step_bar = tqdm(range(self.train_dataloader.__len__()),
                             desc='Train step of epoch %d' % epoch,
                             disable=not is_rank_0())

applications/Chat/coati/trainer/sft.py

@@ -1,7 +1,6 @@
 import math
 import time
-from abc import ABC
+from typing import Optional, List
-from typing import Optional
 
 import loralib as lora
 import torch
@@ -19,11 +18,13 @@ from transformers.trainer import get_scheduler
 
 from colossalai.logging import get_dist_logger
 
+from .callbacks import Callback
+from .base import Trainer
 from .strategies import Strategy
 from .utils import is_rank_0
 
 
-class SFTTrainer(ABC):
+class SFTTrainer(Trainer):
     """
         Trainer to use while training reward model.
 
@@ -35,6 +36,7 @@ class SFTTrainer(ABC):
         eval_dataloader: the dataloader to use for evaluation
         batch_size (int, defaults to 1): the batch size while training
         max_epochs (int, defaults to 2): the number of epochs to train
+        callbacks (List[Callback], defaults to []): the callbacks to call during training process
         optim_kwargs (dict, defaults to {'lr':1e-4}): the kwargs to use while initializing optimizer
     """
 
@@ -48,10 +50,9 @@ class SFTTrainer(ABC):
         batch_size: int = 1,
         max_epochs: int = 2,
         accimulation_steps: int = 8,
+        callbacks: List[Callback] = [],
     ) -> None:
-        super().__init__()
+        super().__init__(strategy, max_epochs, callbacks=callbacks)
-        self.strategy = strategy
-        self.epochs = max_epochs
         self.train_dataloader = train_dataloader
         self.eval_dataloader = eval_dataloader
 
@@ -62,7 +63,7 @@ class SFTTrainer(ABC):
 
         self.accimulation_steps = accimulation_steps
         num_update_steps_per_epoch = len(train_dataloader) // self.accimulation_steps
-        max_steps = math.ceil(self.epochs * num_update_steps_per_epoch)
+        max_steps = math.ceil(self.max_epochs * num_update_steps_per_epoch)
 
         self.scheduler = get_scheduler("cosine",
                                        self.optimizer,
@@ -74,10 +75,10 @@ class SFTTrainer(ABC):
         wandb.watch(self.model)
         total_loss = 0
         # epoch_bar = tqdm(range(self.epochs), desc='Epochs', disable=not is_rank_0())
-        step_bar = tqdm(range(len(self.train_dataloader) // self.accimulation_steps * self.epochs),
+        step_bar = tqdm(range(len(self.train_dataloader) // self.accimulation_steps * self.max_epochs),
                         desc=f'steps',
                         disable=not is_rank_0())
-        for epoch in range(self.epochs):
+        for epoch in range(self.max_epochs):
 
             # process_bar = tqdm(range(len(self.train_dataloader)), desc=f'Train process for{epoch}', disable=not is_rank_0())
             # train
@@ -148,7 +149,7 @@ class SFTTrainer(ABC):
 
                     loss_mean = loss_sum / num_seen
                     if dist.get_rank() == 0:
-                        logger.info(f'Eval Epoch {epoch}/{self.epochs} loss {loss_mean}')
+                        logger.info(f'Eval Epoch {epoch}/{self.max_epochs} loss {loss_mean}')
 
             # epoch_bar.update()
 

applications/Chat/examples/train_dummy.py

@@ -114,8 +114,10 @@ def main(args):
                          eos_token_id=tokenizer.eos_token_id,
                          callbacks=callbacks)
 
-    random_prompts = torch.randint(tokenizer.vocab_size, (1000, 64), device=torch.cuda.current_device())
+    random_prompts = torch.randint(tokenizer.vocab_size, (1000, 1, 64), device=torch.cuda.current_device())
-    trainer.fit(random_prompts,
+    random_attention_mask = torch.randint(1, (1000, 1, 64), device=torch.cuda.current_device()).to(torch.bool)
+    random_pretrain = [{'input_ids':random_prompts[i], 'labels':random_prompts[i], 'attention_mask':random_attention_mask[i]} for i in range(1000)]
+    trainer.fit(random_prompts, random_pretrain,
                 num_episodes=args.num_episodes,
                 max_timesteps=args.max_timesteps,
                 update_timesteps=args.update_timesteps)
@@ -136,7 +138,7 @@ if __name__ == '__main__':
                         default='naive')
     parser.add_argument('--model', type=str, default='gpt2', choices=['gpt2', 'bloom', 'opt', 'roberta'])
     parser.add_argument('--pretrain', type=str, default=None)
-    parser.add_argument('--save_path', type=str, default='actor_checkpoint_dummy.pt')
+    parser.add_argument('--save_path', type=str, default='actor_checkpoint_dummy')
     parser.add_argument('--need_optim_ckpt', type=bool, default=False)
     parser.add_argument('--num_episodes', type=int, default=50)
     parser.add_argument('--max_timesteps', type=int, default=10)

applications/Chat/examples/train_reward_model.py

@@ -3,6 +3,7 @@ from random import randint
 
 import loralib as lora
 import torch
+import torch.distributed as dist
 from coati.dataset import HhRlhfDataset, RmStaticDataset
 from coati.models import LogExpLoss, LogSigLoss
 from coati.models.base import RewardModel
@@ -17,6 +18,8 @@ from coati.trainer.strategies import ColossalAIStrategy, DDPStrategy, NaiveStrat
 from coati.utils import prepare_llama_tokenizer_and_embedding
 from datasets import load_dataset
 from torch.optim import Adam
+from torch.utils.data import DataLoader
+from torch.utils.data.distributed import DistributedSampler
 from transformers import AutoTokenizer, BloomTokenizerFast, DebertaV2Tokenizer, LlamaTokenizer, RobertaTokenizer
 from transformers.models.gpt2.tokenization_gpt2 import GPT2Tokenizer
 
@@ -120,13 +123,38 @@ def train(args):
     else:
         raise ValueError(f'Unsupported dataset "{args.dataset}"')
 
+    if dist.is_initialized() and dist.get_world_size() > 1:
+        train_sampler = DistributedSampler(train_dataset, shuffle=True, seed=42, drop_last=True, rank=dist.get_rank(),
+                                           num_replicas=dist.get_world_size())
+        valid_sampler = DistributedSampler(valid_dataset, shuffle=True, seed=42, drop_last=True, rank=dist.get_rank(),
+                                           num_replicas=dist.get_world_size())
+        eval_sampler = DistributedSampler(eval_dataset, shuffle=True, seed=42, drop_last=True, rank=dist.get_rank(),
+                                          num_replicas=dist.get_world_size())
+    else:
+        train_sampler = None
+        valid_sampler = None
+        eval_sampler = None
+
+    train_dataloader = DataLoader(train_dataset,
+                                  shuffle=(train_sampler is None),
+                                  sampler=train_sampler,
+                                  batch_size=args.batch_size,
+                                  pin_memory=True)
+
+    valid_dataloader = DataLoader(valid_dataset, shuffle=(valid_sampler is None),
+                                  sampler=valid_sampler,
+                                  batch_size=args.batch_size, pin_memory=True)
+
+    eval_dataloader = DataLoader(eval_dataset, shuffle=(eval_sampler is None),
+                                 sampler=eval_sampler, batch_size=args.batch_size, pin_memory=True)
+
     trainer = RewardModelTrainer(model=model,
                                  strategy=strategy,
                                  optim=optim,
                                  loss_fn=loss_fn,
-                                 train_dataset=train_dataset,
+                                 train_dataloader=train_dataloader,
-                                 valid_dataset=valid_dataset,
+                                 valid_dataloader=valid_dataloader,
-                                 eval_dataset=eval_dataset,
+                                 eval_dataloader=eval_dataloader,
                                  batch_size=args.batch_size,
                                  max_epochs=args.max_epochs)