Merge branch 'dev' of https://github.com/RAGEN-AI/vagen into dev

seed: 42

train_path: "./data/sokoban-vision-6-step/train.parquet"

test_path: "./data/sokoban-vision-6-step/test.parquet"

"env1": {

  "env_name": "sokoban",

  'env_config': {

      'num_boxes': 1,

      'dim_room' : [6,6],

      'max_steps': 100,

      'render_mode': "vision",

      'min_actions_to_succeed': 5,

      'max_actions_per_step': 3

  },

  "train_size": 10000,

  "test_size": 1000,

}

 No newline at end of file

        info["metrics"] = metrics

        metrics['turn_metrics']['action_is_effective'] = not np.array_equal(prev_player_position, self.env.player_position)

        self.total_reward += self.reward

        return self._render(init_obs=False), self.reward, done, info

    def system_prompt(self):

# env_utils.py

import os

import random

import logging

import numpy as np