Update README with full model metadata and usage
Browse files
README.md
CHANGED
|
@@ -1,21 +1,72 @@
|
|
| 1 |
---
|
| 2 |
-
language: "es"
|
| 3 |
-
license:
|
| 4 |
tags:
|
| 5 |
- bittensor
|
| 6 |
- subnet-20
|
|
|
|
|
|
|
|
|
|
| 7 |
- tool-calling
|
|
|
|
|
|
|
| 8 |
- agent
|
| 9 |
-
|
| 10 |
-
|
| 11 |
-
datasets:
|
| 12 |
-
- self-generated
|
| 13 |
model-index:
|
| 14 |
-
- name:
|
| 15 |
-
results:
|
| 16 |
-
|
| 17 |
-
|
| 18 |
-
|
|
|
|
|
|
|
|
|
|
| 19 |
---
|
| 20 |
|
| 21 |
-
# Antonio BFCL Toolmodel
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
language: ["es", "en"]
|
| 3 |
+
license: apache-2.0
|
| 4 |
tags:
|
| 5 |
- bittensor
|
| 6 |
- subnet-20
|
| 7 |
+
- bitagent
|
| 8 |
+
- finney
|
| 9 |
+
- tao
|
| 10 |
- tool-calling
|
| 11 |
+
- bfcl
|
| 12 |
+
- reasoning
|
| 13 |
- agent
|
| 14 |
+
base_model: Salesforce/xLAM-7b-r
|
| 15 |
+
pipeline_tag: text-generation
|
|
|
|
|
|
|
| 16 |
model-index:
|
| 17 |
+
- name: antonio-bfcl-toolmodel
|
| 18 |
+
results:
|
| 19 |
+
- task:
|
| 20 |
+
type: text-generation
|
| 21 |
+
name: Generative reasoning and tool-calling
|
| 22 |
+
metrics:
|
| 23 |
+
- type: accuracy
|
| 24 |
+
value: 0.0
|
| 25 |
---
|
| 26 |
|
| 27 |
+
# 馃 Antonio BFCL Toolmodel
|
| 28 |
+
|
| 29 |
+
Este modelo forma parte del ecosistema **BitAgent (Subnet-20)** de Bittensor, dise帽ado para tareas de *tool-calling*, razonamiento l贸gico estructurado y generaci贸n de texto contextual.
|
| 30 |
+
Optimizado para comunicaci贸n eficiente entre agentes dentro del protocolo Finney.
|
| 31 |
+
|
| 32 |
+
---
|
| 33 |
+
|
| 34 |
+
## 馃殌 Descripci贸n t茅cnica
|
| 35 |
+
|
| 36 |
+
**antonio-bfcl-toolmodel** est谩 basado en un modelo open-source tipo `xLAM-7b-r`, ajustado para:
|
| 37 |
+
|
| 38 |
+
- 馃搳 *Razonamiento simb贸lico y factual multiling眉e*
|
| 39 |
+
- 馃З *Tool-calling autom谩tico* (formato JSON conforme a los prompts de Subnet-20)
|
| 40 |
+
- 馃攧 *Respuestas deterministas* con `temperature=0.1` y `top_p=0.9`
|
| 41 |
+
- 鈿欙笍 *Compatibilidad total con el pipeline de BitAgent Miner (v1.0.51)*
|
| 42 |
+
- 馃寪 *Idiomas soportados*: Espa帽ol 馃嚜馃嚫 e Ingl茅s 馃嚞馃嚙
|
| 43 |
+
|
| 44 |
+
---
|
| 45 |
+
|
| 46 |
+
## 馃З Integraci贸n con Subnet-20
|
| 47 |
+
|
| 48 |
+
Los validadores pueden invocar este modelo a trav茅s de los protocolos:
|
| 49 |
+
|
| 50 |
+
- `QueryTask`
|
| 51 |
+
- `QueryResult`
|
| 52 |
+
- `IsAlive`
|
| 53 |
+
- `GetHFModelName`
|
| 54 |
+
- `SetHFModelName`
|
| 55 |
+
|
| 56 |
+
El modelo responde mediante `bittensor.dendrite` y cumple con la especificaci贸n **BitAgent v1.0.51**.
|
| 57 |
+
|
| 58 |
+
---
|
| 59 |
+
|
| 60 |
+
## 馃 Ejemplo de inferencia local
|
| 61 |
+
|
| 62 |
+
```python
|
| 63 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM
|
| 64 |
+
|
| 65 |
+
model_name = "Tonit23/antonio-bfcl-toolmodel"
|
| 66 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
| 67 |
+
model = AutoModelForCausalLM.from_pretrained(model_name)
|
| 68 |
+
|
| 69 |
+
prompt = "Resuelve esta operaci贸n: 12 + 37 = "
|
| 70 |
+
inputs = tokenizer(prompt, return_tensors="pt")
|
| 71 |
+
outputs = model.generate(**inputs, max_new_tokens=32)
|
| 72 |
+
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
|