-
Notifications
You must be signed in to change notification settings - Fork 810
/
Copy pathrun_dapo_qwen2.5_32b.sh
133 lines (122 loc) · 5.45 KB
/
run_dapo_qwen2.5_32b.sh
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
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
#!/usr/bin/env bash
set -euxo pipefail
project_name='DAPO-verl'
exp_name='DAPO-Qwen2.5-32B'
adv_estimator=grpo
use_kl_in_reward=False
kl_coef=0.0
use_kl_loss=False
kl_loss_coef=0.0
clip_ratio_low=0.2
clip_ratio_high=0.28
max_prompt_length=$((1024 * 2))
max_response_length=$((1024 * 20))
enable_overlong_buffer=True
overlong_buffer_len=$((1024 * 4))
overlong_penalty_factor=1.0
loss_agg_mode="token-mean"
enable_filter_groups=True
filter_groups_metric=acc
max_num_gen_batches=10
train_prompt_bsz=512
gen_prompt_bsz=$((train_prompt_bsz * 3))
n_resp_per_prompt=16
train_prompt_mini_bsz=32
# Ray
RAY_ADDRESS=${RAY_ADDRESS:-"http://localhost:8265"}
WORKING_DIR=${WORKING_DIR:-"${PWD}"}
RUNTIME_ENV=${RUNTIME_ENV:-"${WORKING_DIR}/verl/trainer/runtime_env.yaml"}
NNODES=${NNODES:-16}
# Paths
RAY_DATA_HOME=${RAY_DATA_HOME:-"${HOME}/verl"}
MODEL_PATH=${MODEL_PATH:-"${RAY_DATA_HOME}/models/Qwen2.5-32B"}
CKPTS_DIR=${CKPTS_DIR:-"${RAY_DATA_HOME}/ckpts/${project_name}/${exp_name}"}
TRAIN_FILE=${TRAIN_FILE:-"${RAY_DATA_HOME}/data/dapo-math-17k.parquet"}
TEST_FILE=${TEST_FILE:-"${RAY_DATA_HOME}/data/aime-2024.parquet"}
# Algorithm
temperature=1.0
top_p=1.0
top_k=-1 # 0 for HF rollout, -1 for vLLM rollout
val_top_p=0.7
# Performance Related Parameter
sp_size=8
use_dynamic_bsz=True
actor_ppo_max_token_len=$((max_prompt_length + max_response_length))
infer_ppo_max_token_len=$((max_prompt_length + max_response_length))
offload=True
gen_tp=4
ray job submit --no-wait --runtime-env="${RUNTIME_ENV}" \
--working-dir "${WORKING_DIR}" \
-- python3 -m recipe.dapo.src.main_dapo \
data.train_files="${TRAIN_FILE}" \
data.val_files="${TEST_FILE}" \
data.prompt_key=prompt \
data.truncation='left' \
data.max_prompt_length=${max_prompt_length} \
data.max_response_length=${max_response_length} \
data.gen_batch_size=${gen_prompt_bsz} \
data.train_batch_size=${train_prompt_bsz} \
actor_rollout_ref.rollout.n=${n_resp_per_prompt} \
algorithm.adv_estimator=${adv_estimator} \
algorithm.use_kl_in_reward=${use_kl_in_reward} \
algorithm.kl_ctrl.kl_coef=${kl_coef} \
actor_rollout_ref.actor.use_kl_loss=${use_kl_loss} \
actor_rollout_ref.actor.kl_loss_coef=${kl_loss_coef} \
actor_rollout_ref.actor.clip_ratio_low=${clip_ratio_low} \
actor_rollout_ref.actor.clip_ratio_high=${clip_ratio_high} \
actor_rollout_ref.actor.clip_ratio_c=10.0 \
algorithm.filter_groups.enable=${enable_filter_groups} \
algorithm.filter_groups.max_num_gen_batches=${max_num_gen_batches} \
algorithm.filter_groups.metric=${filter_groups_metric} \
actor_rollout_ref.model.use_remove_padding=True \
actor_rollout_ref.actor.use_dynamic_bsz=${use_dynamic_bsz} \
actor_rollout_ref.ref.log_prob_use_dynamic_bsz=${use_dynamic_bsz} \
actor_rollout_ref.rollout.log_prob_use_dynamic_bsz=${use_dynamic_bsz} \
actor_rollout_ref.actor.ppo_max_token_len_per_gpu=${actor_ppo_max_token_len} \
actor_rollout_ref.ref.log_prob_max_token_len_per_gpu=${infer_ppo_max_token_len} \
actor_rollout_ref.rollout.log_prob_max_token_len_per_gpu=${infer_ppo_max_token_len} \
actor_rollout_ref.model.path="${MODEL_PATH}" \
+actor_rollout_ref.model.override_config.attention_dropout=0. \
+actor_rollout_ref.model.override_config.embd_pdrop=0. \
+actor_rollout_ref.model.override_config.resid_pdrop=0. \
actor_rollout_ref.model.enable_gradient_checkpointing=True \
actor_rollout_ref.actor.optim.lr=1e-6 \
actor_rollout_ref.actor.optim.lr_warmup_steps=10 \
actor_rollout_ref.actor.optim.weight_decay=0.1 \
actor_rollout_ref.actor.ppo_mini_batch_size=${train_prompt_mini_bsz} \
actor_rollout_ref.actor.fsdp_config.param_offload=${offload} \
actor_rollout_ref.actor.fsdp_config.optimizer_offload=${offload} \
actor_rollout_ref.actor.entropy_coeff=0 \
actor_rollout_ref.actor.grad_clip=1.0 \
actor_rollout_ref.actor.loss_agg_mode=${loss_agg_mode} \
actor_rollout_ref.actor.ulysses_sequence_parallel_size=${sp_size} \
actor_rollout_ref.rollout.gpu_memory_utilization=0.7 \
actor_rollout_ref.rollout.tensor_model_parallel_size=${gen_tp} \
actor_rollout_ref.rollout.enable_chunked_prefill=True \
actor_rollout_ref.rollout.max_num_batched_tokens=$((max_prompt_length + max_response_length)) \
actor_rollout_ref.rollout.temperature=${temperature} \
actor_rollout_ref.rollout.top_p=${top_p} \
actor_rollout_ref.rollout.top_k="${top_k}" \
actor_rollout_ref.rollout.val_kwargs.temperature=${temperature} \
actor_rollout_ref.rollout.val_kwargs.top_p=${val_top_p} \
actor_rollout_ref.rollout.val_kwargs.top_k=${top_k} \
actor_rollout_ref.rollout.val_kwargs.do_sample=True \
actor_rollout_ref.rollout.val_kwargs.n=1 \
actor_rollout_ref.ref.fsdp_config.param_offload=${offload} \
actor_rollout_ref.ref.ulysses_sequence_parallel_size=${sp_size} \
actor_rollout_ref.actor.fsdp_config.fsdp_size=-1 \
reward_model.reward_manager=dapo \
reward_model.overlong_buffer.enable=${enable_overlong_buffer} \
reward_model.overlong_buffer.len=${overlong_buffer_len} \
reward_model.overlong_buffer.penalty_factor=${overlong_penalty_factor} \
trainer.logger=['console','wandb'] \
trainer.project_name="${project_name}" \
trainer.experiment_name="${exp_name}" \
trainer.n_gpus_per_node=8 \
trainer.nnodes="${NNODES}" \
trainer.val_before_train=True \
trainer.test_freq=5 \
trainer.save_freq=5 \
trainer.total_epochs=1 \
trainer.default_local_dir="${CKPTS_DIR}" \
trainer.resume_mode=auto