open-thought/reasoning-gym
1
0
Fork 0
mirror of https://github.com/open-thought/reasoning-gym.git synced 2026-05-03 17:53:26 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

use native types List->list, Dict->dict, Set->set, Tuple->tuple

Browse source
This commit is contained in:
Andreas Koepf 2025-02-21 15:13:19 +01:00
parent 5d02064b5a
commit 3e7ff3b084
95 changed files with 754 additions and 760 deletions
Download patch file Download diff file Expand all files Collapse all files

6
examples/veRL/main_ppo_custom_reward_server.py
View file

@ -1,7 +1,7 @@
# This example is an adapted version of Bytedance's code:
# https://github.com/volcengine/verl/blob/a65c9157bc0b85b64cd753de19f94e80a11bd871/verl/trainer/main_ppo.py
import os
from typing import Dict, List, Optional
from typing import Optional
import hydra
import ray
@ -64,12 +64,12 @@ class ReasoningGymDataset(Dataset):
self.client.create_experiment(dataset_name, config)
# Cache for batches
self._batch_cache: dict[int, List[BatchEntry]] = {}
self._batch_cache: dict[int, list[BatchEntry]] = {}
def __len__(self) -> int:
return self.size
def _get_batch(self, batch_idx: int) -> List[BatchEntry]:
def _get_batch(self, batch_idx: int) -> list[BatchEntry]:
"""Fetch or retrieve cached batch"""
if batch_idx not in self._batch_cache:
base_index = batch_idx * self.batch_size