Update README.md

README.md

@@ -21,27 +21,68 @@ widget:
 - **Finetuned from model :** llm-jp/llm-jp-3-13b
 
 # Usage
+
+```python
+!pip install unsloth
+!pip install --upgrade --no-cache-dir "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"
+!pip install -U torch
+!pip install -U peft
+```
+
 ```python
+from unsloth import FastLanguageModel
+from peft import PeftModel
 import torch
-from transformers import AutoTokenizer, AutoModelForCausalLM
-tokenizer = AutoTokenizer.from_pretrained("JunichiroMorita/llm-jp-3-13b_lora_20241130")
-model = AutoModelForCausalLM.from_pretrained("JunichiroMorita/llm-jp-3-13b_lora_20241130", device_map="auto", torch_dtype=torch.bfloat16)
-chat = [
-    {"role": "system", "content": "以下は、タスクを説明する指示です。要求を適切に満たす応答を書きなさい。"},
-    {"role": "user", "content": "自然言語処理とは何か"},
-]
-tokenized_input = tokenizer.apply_chat_template(chat, add_generation_prompt=True, tokenize=True, return_tensors="pt").to(model.device)
-with torch.no_grad():
-    output = model.generate(
-        tokenized_input,
-        max_new_tokens=100,
-        do_sample=True,
-        top_p=0.95,
-        temperature=0.7,
-        repetition_penalty=1.05,
-    )[0]
-print(tokenizer.decode(output))
+import json
+from tqdm import tqdm
+import re
+
+model_id = "llm-jp/llm-jp-3-13b"
+adapter_id = f"JunichiroMorita/llm-jp-3-13b-it_lora_20241216"
+
+HF_TOKEN = 'your_hugging_face_token'
+
+dtype = None
+load_in_4bit = True
+
+model, tokenizer = FastLanguageModel.from_pretrained(
+    model_name=model_id,
+    dtype=dtype,
+    load_in_4bit=load_in_4bit,
+    trust_remote_code=True,
+)
+
+model = PeftModel.from_pretrained(model, adapter_id, token=HF_TOKEN)
+
+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 = ""
+
+FastLanguageModel.for_inference(model)
+
+results = []
+for dt in tqdm(datasets):
+  input = dt["input"]
+
+  prompt = f"""### 指示\n{input}\n\n### 回答\n"""
+
+  inputs = tokenizer([prompt], return_tensors = "pt").to(model.device)
+
+  outputs = model.generate(**inputs, max_new_tokens = 512, use_cache = True, do_sample=False, repetition_penalty=1.2)
+  prediction = tokenizer.decode(outputs[0], skip_special_tokens=True).split('\n### 回答\n')[-1]
+
+  results.append({"task_id": dt["task_id"], "input": input, "output": prediction})
 
+with open(f'./llm-jp-3-13b-it_lora_20241216_output.jsonl', 'w', encoding='utf-8') as f:
+    for result in results:
+        json.dump(result, f, ensure_ascii=False)
+        f.write('\n')
 ```
 
 # Data