---
title: gemma-2b-it-function-call
canonical_url: "https://www.modelscope.cn/models/BodhiHu/gemma-2b-it-function-call"
md_url: "https://www.modelscope.cn/models/BodhiHu/gemma-2b-it-function-call.md"
repository: BodhiHu/gemma-2b-it-function-call
last_updated: 2024-04-30
pipeline_tag: question-answering
tasks:
  - question-answering
library_name:
  - transformer
downloads: 13
stars: 0
---

# gemma-2b-it-function-call

> gemma-2b-it-function-call - BodhiHu 在 ModelScope 开源的模型。Model Card for Model ID

- **Repository**: BodhiHu/gemma-2b-it-function-call
- **Tasks**: question-answering
- **Downloads**: 13
- **Stars**: 0
- **Last updated**: 2024-04-30

Source: https://www.modelscope.cn/models/BodhiHu/gemma-2b-it-function-call

---

# Model Card for Model ID

Gemma 2B function calling. [google/gemma-2b-it](https://huggingface.co/google/gemma-2b-it) finetuned on [hypervariance/function-calling-sharegpt](https://huggingface.co/datasets/hypervariance/function-calling-sharegpt).

## Usage

Make sure you have the [peft](https://huggingface.co/docs/peft/en/index) package installed. You can install it with `pip install peft`.


```python
from transformers import AutoModelForCausalLM , AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("bodhicitta/gemma-2b-function-call", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("bodhicitta/gemma-2b-function-call", trust_remote_code=True, device_map="auto")

inputs = tokenizer(prompt,return_tensors="pt").to(model.device)

outputs = model.generate(**inputs,do_sample=True,temperature=0.1,top_p=0.95,max_new_tokens=100)

print(tokenizer.decode(outputs[0]))
```

You can also use sharegpt formatted prompts:

```python
from transformers import AutoModelForCausalLM , AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("bodhicitta/gemma-2b-function-call", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("bodhicitta/gemma-2b-function-call", trust_remote_code=True, device_map="auto")

chat = [
  {
      "from": "system",
      "value": "SYSTEM PROMPT",
  },
  {
      "from": "human",
      "value": "USER QUESTION"
  },
]

prompt = tokenizer.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)

inputs = tokenizer(prompt, return_tensors="pt").to(model.device)

outputs = model.generate(**inputs,do_sample=True,temperature=0.1,top_p=0.95,max_new_tokens=100)

print(tokenizer.decode(outputs[0]))
```

## Prompt template

```text
You are a helpful assistant with access to the following functions. Use them if required -
{
    "name": "function name",
    "description": "function description",
    "parameters": {
        "type": "type (object/number/string)",
        "properties": {
            "property_1": {
                "type": "type",
                "description": "property description"
            }
        },
        "required": [
            "property_1"
        ]
    }
}

To use these functions respond with:
<functioncall> {"name": "function_name", "arguments": {"arg_1": "value_1", "arg_1": "value_1", ...}} </functioncall>

Edge cases you must handle:
 - If there are no functions that match the user request, you will respond politely that you cannot help.

User Question:
USER_QUESTION
```

Function calls are enclosed in `<functioncall>` `</functioncall>`.

The model was trained using the same delimiters as [google/gemma-2b-it](https://huggingface.co/google/gemma-2b-it):

```text
<bos><start_of_turn>user
Write a hello world program<end_of_turn>
<start_of_turn>model
```

Use `<end_of_turn>` stop sequence to prevent the model from generating further text.