essobi commited on
Commit
c5a60f7
·
verified ·
1 Parent(s): 44218e0

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +57 -13
README.md CHANGED
@@ -1,21 +1,65 @@
1
  ---
2
- base_model: unsloth/functiongemma-270m-it
 
3
  tags:
4
- - text-generation-inference
5
- - transformers
6
- - unsloth
7
- - gemma3_text
8
- license: apache-2.0
 
 
 
 
 
9
  language:
10
- - en
11
  ---
12
 
13
- # Uploaded finetuned model
14
 
15
- - **Developed by:** essobi
16
- - **License:** apache-2.0
17
- - **Finetuned from model :** unsloth/functiongemma-270m-it
18
 
19
- This gemma3_text model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
20
 
21
- [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ license: gemma
3
+ library_name: transformers
4
  tags:
5
+ - function-calling
6
+ - tool-use
7
+ - mobile
8
+ - gemma
9
+ - unsloth
10
+ - fine-tuned
11
+ base_model: google/gemma-3-1b-it
12
+ datasets:
13
+ - google/mobile-actions
14
+ pipeline_tag: text-generation
15
  language:
16
+ - en
17
  ---
18
 
19
+ # FunctionGemma Mobile Actions v6
20
 
21
+ A fine-tuned version of [FunctionGemma 270M](https://huggingface.co/google/gemma-3-1b-it) optimized for mobile device function calling. This model excels at understanding natural language commands and mapping them to structured function calls for common mobile actions.
 
 
22
 
23
+ ## Model Description
24
 
25
+ - **Base Model:** google/gemma-3-1b-it (270M parameters)
26
+ - **Fine-tuning Method:** LoRA (r=128, alpha=128)
27
+ - **Training Data:** [google/mobile-actions](https://huggingface.co/datasets/google/mobile-actions) + synthetic augmentation
28
+ - **Optimized For:** Mobile assistant function calling
29
+
30
+ ## Supported Functions
31
+
32
+ | Function | Description | Example Input |
33
+ |----------|-------------|---------------|
34
+ | `set_alarm` | Set alarms | "Wake me up at 7am" |
35
+ | `create_reminder` | Create reminders | "Remind me to buy milk" |
36
+ | `set_timer` | Set countdown timers | "Timer for 10 minutes" |
37
+ | `make_call` | Make phone calls | "Call Mom" |
38
+ | `send_message` | Send text messages | "Text John I'm running late" |
39
+ | `create_calendar_event` | Schedule events | "Schedule meeting at 3pm" |
40
+ | `play_music` | Play music | "Play some jazz" |
41
+ | `get_weather` | Get weather info | "What's the weather like?" |
42
+ | `open_app` | Open applications | "Open the camera" |
43
+ | `navigate` | Get directions | "Navigate to the airport" |
44
+ | `set_volume` | Adjust volume | "Turn the volume up" |
45
+ | `calculator` | Math calculations | "What's 15 times 23?" |
46
+
47
+ ## Usage
48
+
49
+ ```python
50
+ from vllm import LLM, SamplingParams
51
+
52
+ llm = LLM(model="essobi/functiongemma-mobile-actions-v6-16bit", trust_remote_code=True)
53
+
54
+ # See full documentation for prompt format and tool definitions
55
+ ```
56
+
57
+ ## Output Format
58
+
59
+ ```
60
+ <start_function_call>call:function_name{param1:<escape>value1<escape>,param2:<escape>value2<escape>}<end_function_call>
61
+ ```
62
+
63
+ ## License
64
+
65
+ This model is released under the [Gemma License](https://ai.google.dev/gemma/terms).