---
datasets:
- kinokokoro/ichikara-instruction-003
language:
- ja
base_model:
- llm-jp/llm-jp-3-13b
---

elyza-tasks-100-TV_0.jsonl の回答モデルの作成のためのコードです。
サンプルコードに対して以下の変更を行いスコア改善を試みました。
- データセットを ichikara-instruction-003 の全てのファイルを利用するよう変更
- 学習率(learning_rate) を 2e-5へ変更
- 累積勾配(gradient_accumulation_steps) を 4 に変更
- RoRAのRANK(LoraConfig r)を 32 に変更

自宅のPC(RTX3090) でコードを実行し、解答を出力しました。

```python

import wandb
import os

WANDB_API_KEY = "my-token"
wandb.login(key=WANDB_API_KEY)
wandb.init(project='llm2024-competition')

HF_TOKEN = "my-token"

from transformers import (
    AutoModelForCausalLM,
    AutoTokenizer,
    BitsAndBytesConfig,
    TrainingArguments,
    logging,
)
from peft import (
    LoraConfig,
    PeftModel,
    get_peft_model,
)
import os, torch, gc
from datasets import load_dataset
import bitsandbytes as bnb
from trl import SFTTrainer

SEED_VALUE = 42

base_model_id = "llm-jp/llm-jp-3-13b" 
new_model_id = "llm-jp-3-13b-finetune" #Fine-Tuningしたモデルにつけたい名前

bnb_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4", # nf4は通常のINT4より精度が高く、ニューラルネットワークの分布に最適です
    bnb_4bit_compute_dtype=torch.bfloat16,
)


model = AutoModelForCausalLM.from_pretrained(
    base_model_id,
    quantization_config=bnb_config,
    device_map="cuda:0" #auto"
)

tokenizer = AutoTokenizer.from_pretrained(base_model_id, trust_remote_code=True)


def find_all_linear_names(model):
    cls = bnb.nn.Linear4bit # 4bit量子化線形層クラスを指定
    lora_module_names = set() # ここに取得した線形層を保持します。

    # モデル内の全てのモジュールを探索します
    for name, module in model.named_modules():
        if isinstance(module, cls): # モジュールが4bit量子化線形層の場合
            names = name.split('.') # モジュールの名前を分割 (ネストされてる際などに対処)
            lora_module_names.add(names[0] if len(names) == 1 else names[-1]) # 最下層の名前をlora_module_namesに追加

    # 'lm_head' は16ビット演算の際に除外する必要があるため、lora_module_namesから削除
    if 'lm_head' in lora_module_names:
        lora_module_names.remove('lm_head')

    return list(lora_module_names) # lora_module_namesをリストに変換して返します。

modules = find_all_linear_names(model)

peft_config = LoraConfig(
    r=32,   #16,
    lora_alpha=32,
    lora_dropout=0.05,
    bias="none",
    task_type="CAUSAL_LM",
    target_modules=modules,
)

model = get_peft_model(model, peft_config)


from datasets import concatenate_datasets, DatasetDict

# 全てのデータセットを読み込み
dataset0 = load_dataset("json", data_files="./Distribution20241221_all/ichikara-instruction-003-001-1.json")
dataset1 = load_dataset("json", data_files="./Distribution20241221_all/ichikara-instruction-003-001-1.json")
dataset2 = load_dataset("json", data_files="./Distribution20241221_all/ichikara-instruction-003-001-2.2.json")
dataset3 = load_dataset("json", data_files="./Distribution20241221_all/ichikara-instruction-003-001-5.2.json")
dataset4 = load_dataset("json", data_files="./Distribution20241221_all/ichikara-instruction-003-001-2.1.json")
dataset5 = load_dataset("json", data_files="./Distribution20241221_all/ichikara-instruction-003-001-5.1.json")
dataset6 = load_dataset("json", data_files="./Distribution20241221_all/ichikara-instruction-003-002-1.json")
dataset7 = load_dataset("json", data_files="./Distribution20241221_all/ichikara-instruction-003-003-1.json")

datasets_to_concatenate = [
    dataset0["train"], 
    dataset1["train"], 
    dataset2["train"],
    dataset3["train"],
    dataset4["train"],
    dataset5["train"],
    dataset6["train"],
    dataset7["train"]
    ]

concatenated_train_dataset = concatenate_datasets(datasets_to_concatenate)

dataset_all = DatasetDict({
    "train": concatenated_train_dataset
})

# 結合したデータを使用
dataset=dataset_all

# 学習時のプロンプトフォーマットの定義
prompt = """### 指示
{}
### 回答
{}"""


"""
formatting_prompts_func: 各データをプロンプトに合わせた形式に合わせる
"""
EOS_TOKEN = tokenizer.eos_token # トークナイザーのEOSトークン（文末トークン）
def formatting_prompts_func(examples):
    input = examples["text"] # 入力データ
    output = examples["output"] # 出力データ
    text = prompt.format(input, output) + EOS_TOKEN # プロンプトの作成
    return { "formatted_text" : text, } # 新しいフィールド "formatted_text" を返す
pass

# # 各データにフォーマットを適用
dataset = dataset.map(
    formatting_prompts_func,
    num_proc= 4, # 並列処理数を指定
)

# データをtrainデータとtestデータに分割 (test_sizeの比率に)
dataset = dataset["train"].train_test_split(test_size=0.1, seed=SEED_VALUE)


training_arguments = TrainingArguments(
    output_dir=new_model_id,
    per_device_train_batch_size=1,  #
    gradient_accumulation_steps=4,  # def: 2
    optim="paged_adamw_32bit",
    num_train_epochs=1,            # def: 1
    logging_strategy="steps",
    logging_steps=10,
    warmup_steps=10,
    save_steps=100,
    save_total_limit = 2,
    max_steps = -1,                 # def:-1
    learning_rate=2e-5,             # def:5e-5,
    fp16= False,
    bf16= False,
    seed = SEED_VALUE,
    group_by_length=True,
    report_to="wandb"
)

trainer = SFTTrainer(
    model=model,
    train_dataset=dataset["train"],
    peft_config=peft_config,
    max_seq_length= 512,
    dataset_text_field="formatted_text",
    tokenizer=tokenizer,
    args=training_arguments,
    packing= False,
)

model.config.use_cache = False # キャッシュ機能を無効化
trainer.train() # トレーニングを実行

from datetime import datetime

# 現在の日時を取得
now = datetime.now()

# フォーマットを指定して日時を文字列に変換
formatted_date = now.strftime("%Y%m%d_%H%M%S")  # 例: "20241214_153045"

print(formatted_date)

# タスクとなるデータの読み込み。
# 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。
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 = ""


# モデルによるタスクの推論。
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})

# こちらで生成されたjsolを提出してください。
# 本コードではinputとeval_aspectも含んでいますが、なくても問題ありません。
# 必須なのはtask_idとoutputとなります。
import re
jsonl_id = re.sub(".*/", "", new_model_id)
with open(f"./{jsonl_id}-outputs-{formatted_date}.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')

# モデルとトークナイザーをHugging Faceにアップロード
model.push_to_hub(new_model_id, token=HF_TOKEN, private=True) # Online saving
tokenizer.push_to_hub(new_model_id, token=HF_TOKEN, private=True) # Online saving



```



---
library_name: transformers
---

# Model Card for Model ID

<!-- Provide a quick summary of what the model is/does. -->

## Model Details

### Model Description

<!-- Provide a longer summary of what this model is. -->

This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.

- **Developed by:** [More Information Needed]
- **Funded by [optional]:** [More Information Needed]
- **Shared by [optional]:** [More Information Needed]
- **Model type:** [More Information Needed]
- **Language(s) (NLP):** [More Information Needed]
- **License:** [More Information Needed]
- **Finetuned from model [optional]:** [More Information Needed]

### Model Sources [optional]

<!-- Provide the basic links for the model. -->

- **Repository:** [More Information Needed]
- **Paper [optional]:** [More Information Needed]
- **Demo [optional]:** [More Information Needed]

## Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->

### Direct Use

<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->

[More Information Needed]

### Downstream Use [optional]

<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->

[More Information Needed]

### Out-of-Scope Use

<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->

[More Information Needed]

## Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->

[More Information Needed]

### Recommendations

<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

## How to Get Started with the Model

Use the code below to get started with the model.

[More Information Needed]

## Training Details

### Training Data

<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->

[More Information Needed]

### Training Procedure

<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->

#### Preprocessing [optional]

[More Information Needed]


#### Training Hyperparameters

- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->

#### Speeds, Sizes, Times [optional]

<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->

[More Information Needed]

## Evaluation

<!-- This section describes the evaluation protocols and provides the results. -->

### Testing Data, Factors & Metrics

#### Testing Data

<!-- This should link to a Dataset Card if possible. -->

[More Information Needed]

#### Factors

<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->

[More Information Needed]

#### Metrics

<!-- These are the evaluation metrics being used, ideally with a description of why. -->

[More Information Needed]

### Results

[More Information Needed]

#### Summary



## Model Examination [optional]

<!-- Relevant interpretability work for the model goes here -->

[More Information Needed]

## Environmental Impact

<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->

Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).

- **Hardware Type:** [More Information Needed]
- **Hours used:** [More Information Needed]
- **Cloud Provider:** [More Information Needed]
- **Compute Region:** [More Information Needed]
- **Carbon Emitted:** [More Information Needed]

## Technical Specifications [optional]

### Model Architecture and Objective

[More Information Needed]

### Compute Infrastructure

[More Information Needed]

#### Hardware

[More Information Needed]

#### Software

[More Information Needed]

## Citation [optional]

<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->

**BibTeX:**

[More Information Needed]

**APA:**

[More Information Needed]

## Glossary [optional]

<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->

[More Information Needed]

## More Information [optional]

[More Information Needed]

## Model Card Authors [optional]

[More Information Needed]

## Model Card Contact

[More Information Needed]