Qwen2-7B-Instruct-glaive-function-calling

Qwen2-7B-Instruct-glaive-function-calling

Introduction

基于数据集 glaive-function-calling-v2 在 Qwen2-7B-Instruct 上进行微调而来 <br>

Training details

使用 lora 进行训练 训练样本如下：

<|im_start|>system
You are a helpful assistant with access to the following functions. Use them if required -
{
    "name": "generate_invoice",
    "description": "Generate an invoice with specified details",
    "parameters": {
        "type": "object",
        "properties": {
            "customer_name": {
                "type": "string",
                "description": "The name of the customer"
            },
            "items": {
                "type": "array",
                "items": {
                    "type": "object",
                    "properties": {
                        "name": {
                            "type": "string",
                            "description": "The name of the item"
                        },
                        "quantity": {
                            "type": "integer",
                            "description": "The quantity of the item"
                        },
                        "price": {
                            "type": "number",
                            "description": "The price of the item"
                        }
                    },
                    "required": [
                        "name",
                        "quantity",
                        "price"
                    ]
                }
            }
        },
        "required": [
            "customer_name",
            "items"
        ]
    }
}
<|im_end|>
<|im_start|>user
I need to generate an invoice for a customer named John Doe. He bought 2 apples for $1 each and 3 oranges for $2 each.<|im_end|>
<|im_start|>assistant
<functioncall> {"name": "generate_invoice", "arguments": '{"customer_name": "John Doe", "items": [{"name": "apple", "quantity": 2, "price": 1}, {"name": "orange", "quantity": 3, "price": 2}]}'} <|endoftext|><|im_end|>
<|im_start|>function
{"invoice_id": "INV12345", "customer_name": "John Doe", "items": [{"name": "apple", "quantity": 2, "price": 1, "total": 2}, {"name": "orange", "quantity": 3, "price": 2, "total": 6}], "total": 8, "status": "Generated"}<|im_end|>
<|im_start|>assistant
The invoice has been successfully generated. The invoice ID is INV12345. The total amount for 2 apples and 3 oranges is $8. <|endoftext|><|im_end|>

Quickstart

参考 Qwen2-7B-Instruct

Here provides a code snippet with apply_chat_template to show you how to load the tokenizer and model and how to generate contents.

from transformers import AutoModelForCausalLM, AutoTokenizer
device = "cuda" # the device to load the model onto

model = AutoModelForCausalLM.from_pretrained(
    "Qwen/Qwen2-7B-Instruct",
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2-7B-Instruct")

prompt = "Give me a short introduction to large language model."
messages = [
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True
)
model_inputs = tokenizer([text], return_tensors="pt").to(device)

generated_ids = model.generate(
    model_inputs.input_ids,
    max_new_tokens=512
)
generated_ids = [
    output_ids[len(input_ids):] for input_ids, output_ids in zip(model_inputs.input_ids, generated_ids)
]

response = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]