File size: 1,872 Bytes
0185029 | 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 | # Copyright 2024 Bytedance Ltd. and/or its affiliates
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""
ActorRolloutRef config
"""
from dataclasses import dataclass, field
from .actor import ActorConfig, FSDPConfig, ModelConfig, OptimConfig, RefConfig
from .critic import CriticConfig
from .reward import RewardConfig
from .rollout import RolloutConfig
__all__ = [
"ActorConfig",
"CriticConfig",
"FSDPConfig",
"ModelConfig",
"OptimConfig",
"RefConfig",
"RewardConfig",
"RolloutConfig",
"WorkerConfig",
]
@dataclass
class WorkerConfig:
hybrid_engine: bool = True
actor: ActorConfig = field(default_factory=ActorConfig)
critic: CriticConfig = field(default_factory=CriticConfig)
ref: RefConfig = field(default_factory=RefConfig)
reward: RewardConfig = field(default_factory=RewardConfig)
rollout: RolloutConfig = field(default_factory=RolloutConfig)
def post_init(self):
self.ref.micro_batch_size_per_device_for_experience = self.actor.micro_batch_size_per_device_for_experience
self.ref.padding_free = self.actor.padding_free
self.ref.dynamic_batching = self.actor.dynamic_batching
self.ref.max_token_len_per_gpu = self.actor.max_token_len_per_gpu
self.ref.ulysses_size = self.actor.ulysses_size
self.ref.use_torch_compile = self.actor.use_torch_compile
|