!pip install -U bitsandbytes
!pip install -U transformers
!pip install -U accelerate
!pip install -U datasets
!pip install ipywidgets --upgrade
from transformers import (
    AutoModelForCausalLM,
    AutoTokenizer,
    BitsAndBytesConfig,
)
import torch
from tqdm import tqdm
import json

HF_TOKEN = "<my token>"

model_name = "papettoTV/llm-jp-3-13b-it"

# QLoRA config
bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.bfloat16,
    bnb_4bit_use_double_quant=False,
)

# Load model
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    quantization_config=bnb_config,
    device_map="auto",
    token = HF_TOKEN
)

# Load tokenizer
tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True, token = HF_TOKEN)

# データセットの読み込み。
# omnicampusの開発環境では、左にタスクのjsonlをドラッグアンドドロップしてから実行。
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 = ""

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

  input = data["input"]
  prompt = f"""### 指示
  {input}
  ### 回答：
  """

  input_ids = tokenizer(prompt, return_tensors="pt").to(model.device)
  # outputs = model.generate(**input_ids, max_new_tokens=512, do_sample=False, repetition_penalty=1.2,)
  outputs = model.generate(input_ids=input_ids.input_ids, max_new_tokens=512, do_sample=False, repetition_penalty=1.2,)  

  output = tokenizer.decode(outputs[0][input_ids.input_ids.size(1):], skip_special_tokens=True)

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

# llmjp
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)
  with torch.no_grad():
      outputs = model.generate(
          tokenized_input,
          max_new_tokens=100,
          do_sample=False,
          repetition_penalty=1.2
      )[0]
  output = tokenizer.decode(outputs[tokenized_input.size(1):], skip_special_tokens=True)

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

# こちらで生成されたjsolを提出しました
import re
model_name = re.sub(".*/", "", model_name)
with open(f"./{model_name}-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')