inspo/scripts/train_baseline.sh at main · cambridgeltl/inspo · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
#!/bin/bash

set -euo pipefail
mkdir -p logs

DATA_NAME=nq_hotpotqa_train
export DATA_DIR=data/${DATA_NAME}
export BASE_MODEL='Qwen/Qwen2.5-3B'
export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
export VLLM_ATTENTION_BACKEND=XFORMERS

WANDB_PROJECT="INSPO"
EXPERIMENT_NAME="baseline_${DATA_NAME}_$(basename ${BASE_MODEL})"

PYTHONUNBUFFERED=1 python3 -m inspo.train \
    data.train_files=$DATA_DIR/train.parquet \
    data.val_files=$DATA_DIR/test.parquet \
    data.train_data_num=null \
    data.val_data_num=null \
    data.train_batch_size=512 \
    data.val_batch_size=256 \
    data.max_prompt_length=4096 \
    data.max_response_length=500 \
    data.max_start_length=2048 \
    data.max_obs_length=500 \
    data.shuffle_train_dataloader=True \
    algorithm.adv_estimator=grpo \
    algorithm.no_think_rl=false \
    actor_rollout_ref.model.path=$BASE_MODEL \
    actor_rollout_ref.model.enable_gradient_checkpointing=true \
    actor_rollout_ref.model.use_remove_padding=True \
    actor_rollout_ref.actor.optim.lr=5e-7 \
    actor_rollout_ref.actor.optim.lr_warmup_steps_ratio=0.1 \
    actor_rollout_ref.actor.optim.warmup_style=cosine \
    actor_rollout_ref.actor.use_kl_loss=true \
    actor_rollout_ref.actor.kl_loss_coef=0.001 \
    actor_rollout_ref.actor.kl_loss_type=low_var_kl \
    actor_rollout_ref.actor.ppo_mini_batch_size=256 \
    actor_rollout_ref.actor.ppo_micro_batch_size=64 \
    actor_rollout_ref.actor.state_masking=true \
    actor_rollout_ref.actor.fsdp_config.param_offload=true \
    actor_rollout_ref.actor.fsdp_config.grad_offload=true \
    actor_rollout_ref.actor.fsdp_config.optimizer_offload=true \
    actor_rollout_ref.rollout.name=vllm \
    actor_rollout_ref.rollout.tensor_model_parallel_size=1 \
    actor_rollout_ref.rollout.gpu_memory_utilization=0.6 \
    actor_rollout_ref.rollout.log_prob_micro_batch_size=128 \
    actor_rollout_ref.rollout.n_agent=5 \
    actor_rollout_ref.rollout.temperature=1 \
    actor_rollout_ref.ref.log_prob_micro_batch_size=128 \
    actor_rollout_ref.ref.fsdp_config.param_offload=True \
    trainer.logger=['wandb'] \
    +trainer.val_only=false \
    +trainer.val_before_train=false \
    trainer.default_hdfs_dir=null \
    trainer.n_gpus_per_node=8 \
    trainer.nnodes=1 \
    trainer.save_freq=400 \
    trainer.test_freq=100 \
    trainer.project_name=$WANDB_PROJECT \
    trainer.experiment_name=$EXPERIMENT_NAME \
    trainer.total_epochs=15 \
    trainer.total_training_steps=305 \
    trainer.default_local_dir=verl_checkpoints/$EXPERIMENT_NAME \
    max_turns=4 \
    retriever.url="http://127.0.0.1:8000/retrieve" \
    retriever.topk=3 \
    inspo.enable=false \
    debug_mode=false \
    2>&1 | tee logs/${EXPERIMENT_NAME}.log