File tree 5 files changed +5
-5
lines changed 5 files changed +5
-5
lines changed Original file line number Diff line number Diff line change 1
1
log_each_n_secs : 10
2
2
llm_max_rollouts : 128
3
3
rollout_workers : 1
4
- rollout_policy : pipelinerl.math_rollouts .generate_math_rollout
4
+ rollout_policy : pipelinerl.math.rollouts .generate_math_rollout
5
5
discount_factor : 1
6
6
system_prompt : Please reason step by step, and put your final answer within \boxed{}.
7
7
task_template : |-
Original file line number Diff line number Diff line change 1
1
import hydra
2
2
from omegaconf import DictConfig
3
3
4
- from pipelinerl .verifier_api import run_verifier
4
+ from pipelinerl .math . verifier_api import run_verifier
5
5
from pipelinerl .utils import better_crashing
6
6
7
7
Original file line number Diff line number Diff line change 8
8
9
9
from pipelinerl .async_llm import llm_async_generate
10
10
from pipelinerl .finetune .data import MASKED_TOKEN_ID
11
- from pipelinerl .verifier_api import verify_answer_rpc
11
+ from pipelinerl .math . verifier_api import verify_answer_rpc
12
12
13
13
14
14
class RewardTable (BaseModel ):
File renamed without changes.
Original file line number Diff line number Diff line change 20
20
import wandb
21
21
from pipelinerl .finetune .logging_ import flatten_dict_config , init_wandb
22
22
from pipelinerl .load_datasets import load_datasets
23
- from pipelinerl .math_rollouts import RolloutResult
23
+ from pipelinerl .math . rollouts import RolloutResult
24
24
from pipelinerl .shared_memory_array import SharedMemoryArray
25
25
from pipelinerl .state import TrainerState
26
26
from pipelinerl .streams import (
30
30
set_streams_backend ,
31
31
write_to_streams ,
32
32
)
33
- from pipelinerl .verifier_api import wait_for_verifier
33
+ from pipelinerl .math . verifier_api import wait_for_verifier
34
34
35
35
from .utils import (
36
36
always_or_never_success_stats ,
You can’t perform that action at this time.
0 commit comments