CoolFace
Modelpublic

xczou/distilbert-intent-sql-financial-general

sourceHugging Faceapache-2.0updated 5mo agoView on Hugging Face
0likes8downloads
Model Card

distilbert-intent-sql-financial-general

Fine-tuned distilbert-base-uncased for 3-class intent routing in an LLM inference pipeline.

Purpose

Routes user prompts to the appropriate vLLM LoRA adapter on a Triton Inference Server:

LabelIDRoutes to
GENERAL0Qwen2.5-7B-Instruct (no LoRA)
SQL1sql-expert LoRA adapter
FINANCIAL2financial LoRA adapter

Training

  • —Base model: distilbert/distilbert-base-uncased
  • —Dataset: 91 hand-curated examples (SQL=30, FINANCIAL=30, GENERAL=31)
  • —Epochs: 5
  • —Learning rate: 2e-5
  • —Batch size: 16
  • —Max sequence length: 128
  • —Optimizer: AdamW (weight_decay=0.01)
  • —Val split: 20% stratified

Deployment

Exported to ONNX (opset 17) via optimum and served as an ONNX Runtime backend model inside NVIDIA Triton Inference Server on GKE Autopilot (NVIDIA L4 GPU).

Usage

python
from transformers import pipeline

classifier = pipeline("text-classification", model="xczou/distilbert-intent-sql-financial-general")
classifier("Write a SQL query to find all orders above 100")
# [{'label': 'SQL', 'score': 0.98}]