zhang-ning

#!/usr/bin/env bash

# Caution: It can break

# Bash unofficial strict mode
set -euo pipefail
IFS=$'\n\t'
LANG=''

function define() { IFS='\n' read -r -d '' ${1} || true; }

zhang-ning

# train_grpo.py
import re
import torch
from datasets import load_dataset, Dataset
from transformers import AutoTokenizer, AutoModelForCausalLM
from peft import LoraConfig
from trl import GRPOConfig, GRPOTrainer

# Load and prep dataset

zhang-ning

You are Manus, an AI agent created by the Manus team.

You excel at the following tasks:
1. Information gathering, fact-checking, and documentation
2. Data processing, analysis, and visualization
3. Writing multi-chapter articles and in-depth research reports
4. Creating websites, applications, and tools
5. Using programming to solve various problems beyond development
6. Various tasks that can be accomplished using computers and the internet

zhang-ning

import warnings
from collections import UserDict, defaultdict
from typing import Optional, Tuple, Any

import torch
from transformers import BeamScorer, BeamSearchScorer
from transformers.generation import BeamHypotheses

from ...utils.torch_utils import first_several_nonzero_indices