from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
from peft import PeftModel
import torch

# ベースモデル名とアダプターのリポジトリ名
base_model_name = "llm-jp/llm-jp-3-13b"  # 事前学習済みモデル
revision = "cd3823f4c1fcbb0ad2e2af46036ab1b0ca13192a"
adapter_repo_id = "Eito2002/llm-jp-3-13b-finetune"

# QLoRAの設定
bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.bfloat16,
)

# トークナイザーとベースモデルを読み込み
tokenizer = AutoTokenizer.from_pretrained(base_model_name, revision=revision)
base_model = AutoModelForCausalLM.from_pretrained(base_model_name, revision=revision, quantization_config=bnb_config, device_map="auto")

# Hugging Faceからアダプターを読み込み
model = PeftModel.from_pretrained(base_model, adapter_repo_id)

# モデルの確認
print("アダプターが統合されました！")
print(model)

# 推論テキスト
prompt = "AIとは何ですか？"

# トークナイズ
tokenized_input = tokenizer.encode(prompt, add_special_tokens=False, return_tensors="pt").to(model.device)
attention_mask = torch.ones_like(tokenized_input)

# 推論を実行
with torch.no_grad():
    outputs = model.generate(
        tokenized_input,
        attention_mask=attention_mask,
        max_new_tokens=100,
        do_sample=False,
        repetition_penalty=1.2,
        pad_token_id=tokenizer.eos_token_id
    )[0]
output = tokenizer.decode(outputs[tokenized_input.size(1):], skip_special_tokens=True)

print(output)

import json

datasets = []
with open("./elyza-tasks-100-TV_0.jsonl", "r") as f:
    item = ""
    for line in f:
      line = line.strip()
      item += line
      if item.endswith("}"):
        datasets.append(json.loads(item))
        item = ""

from tqdm import tqdm

results = []
for data in tqdm(datasets):

  input = data["input"]

  prompt = f"""### 指示
  {input}
  ### 回答
  """

  tokenized_input = tokenizer.encode(prompt, add_special_tokens=False, return_tensors="pt").to(model.device)
  attention_mask = torch.ones_like(tokenized_input)
  with torch.no_grad():
      outputs = model.generate(
          tokenized_input,
          attention_mask=attention_mask,
          max_new_tokens=100,
          do_sample=False,
          repetition_penalty=1.2,
          pad_token_id=tokenizer.eos_token_id
      )[0]
  output = tokenizer.decode(outputs[tokenized_input.size(1):], skip_special_tokens=True)

  results.append({"task_id": data["task_id"], "input": input, "output": output})

with open(f"./elyza-tasks-100-TV-outputs.jsonl", 'w', encoding='utf-8') as f:
    for result in results:
        json.dump(result, f, ensure_ascii=False)  # ensure_ascii=False for handling non-ASCII characters
        f.write('\n')