Downloads · 30 days
0
tmdoi/llm-jp-3-13b-ft-CHGPara-myds-20241216c-ep1
llm-jp-3-13b-ft-CHGPara-myds-20241216c-ep1 is a machine learning model from tmdoi. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
以下の手順に従うことで、Hugging Face上のモデル(llm-jp/llm-jp-3-13b + tmdoi/llm-jp-3-13b-ft-CHGPara-myds-20241216c-ep1)を用いて入力データ(elyza-tasks-100-TV0.jsonl)を推論し、その結果を{adapterid}-outputs.jsonlというファイルに出力できます。
Downloads · 30 days
0
Access
Public
Updated Dec 26, 2024
Repo size
157 MB
Likes
1
Public
Click a slice to open those files.
.safetensors157 MB · 96%
From the Hugging Face model README
以下の手順に従うことで、Hugging Face上のモデル(llm-jp/llm-jp-3-13b + tmdoi/llm-jp-3-13b-ft-CHGPara-myds-20241216c-ep1)を用いて入力データ(elyza-tasks-100-TV_0.jsonl)を推論し、その結果を{adapter_id}-outputs.jsonlというファイルに出力できます。
!pip install -U bitsandbytes
!pip install -U transformers
!pip install -U accelerate
!pip install -U datasets
!pip install -U peft
!pip install ipywidgets --upgrade
from transformers import (
AutoModelForCausalLM,
AutoTokenizer,
BitsAndBytesConfig,
)
from peft import PeftModel
import torch
from tqdm import tqdm
import json
# Hugging Faceで取得したToken
HF_TOKEN = "Hugging Face Token"
# ベースとなるモデルと学習したLoRAのアダプタ
model_id = "llm-jp/llm-jp-3-13b"
adapter_id = "tmdoi/llm-jp-3-13b-ft-CHGPara-myds-20241215f-ep1"
# QLoRA config
bnb_config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_quant_type="nf4",
bnb_4bit_compute_dtype=torch.bfloat16,
)
# Load model
model = AutoModelForCausalLM.from_pretrained(
model_id,
quantization_config=bnb_config,
device_map="auto",
token = HF_TOKEN
)
# Load tokenizer
tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True, token = HF_TOKEN)
# 元のモデルにLoRAのアダプタを統合
model = PeftModel.from_pretrained(model, adapter_id, token = HF_TOKEN)
# データセットの読み込み。
datasets = []
# with open("./elyza-tasks-100-TV_0.jsonl", "r") as f:
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 = ""
# 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:
with open("/content/drive/MyDrive/SFT_outputs/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 = ""
# 推論実行
# llmjp
print(adapter_id)
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=512,
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)
#目視確認用の出力
print("==INPUT===")
print(input)
print("==OUTPUT===")
print(output)
results.append({"task_id": data["task_id"], "input": input, "output": output})
## 出力の保存
import re
jsonl_id = re.sub(".*/", "", adapter_id)
with open(f"{output_GD_dir}/{new_model_id}/{jsonl_id}-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')