NetworkAgent2 FunctionGemma

Fine-tuning de google/functiongemma-270m-it para chamadas de fun??o de diagn?stico de rede no app Android NetworkAgent2.

Ferramentas treinadas

get_number_connected_users: n?mero de dispositivos conectados.
check_internet_status: status da conex?o com a internet.
get_internet_speed: velocidade atual da internet em Mbps.

Dataset

O treino usa o dataset lucasllfsQ/network-agent-functiongemma-en-es. Uma c?pia dos splits usados ? enviada em dataset/.

Artefatos

adapter/: adapter LoRA/PEFT.
merged/: modelo Hugging Face ap?s merge do LoRA.
litert/: export LiteRT/LiteRT-LM ou instru??es de convers?o geradas pelo notebook.

Uso r?pido

from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "lucasllfsQ/networkagent2-functiongemma-litert"
tokenizer = AutoTokenizer.from_pretrained(model_id, subfolder="merged")
model = AutoModelForCausalLM.from_pretrained(model_id, subfolder="merged", device_map="auto")

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for lucasllfsQ/networkagent2-functiongemma-litert

Base model

google/functiongemma-270m-it

Finetuned

(429)

this model