Text Classification
Transformers
Safetensors
PEFT
English
roberta
nlp
lora
multitask-learning
customer-support
text-embeddings-inference
Instructions to use San-Analytics/TicketIQ-MultiTask with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use San-Analytics/TicketIQ-MultiTask with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="San-Analytics/TicketIQ-MultiTask")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("San-Analytics/TicketIQ-MultiTask") model = AutoModelForSequenceClassification.from_pretrained("San-Analytics/TicketIQ-MultiTask", device_map="auto") - PEFT
How to use San-Analytics/TicketIQ-MultiTask with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: en
|
| 3 |
+
license: mit
|
| 4 |
+
pipeline_tag: text-classification
|
| 5 |
+
tags:
|
| 6 |
+
- nlp
|
| 7 |
+
- transformers
|
| 8 |
+
- roberta
|
| 9 |
+
- peft
|
| 10 |
+
- lora
|
| 11 |
+
- multitask-learning
|
| 12 |
+
- customer-support
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
# TicketIQ-MultiTask
|
| 16 |
+
|
| 17 |
+
TicketIQ is a multi-task NLP model for automated customer support ticket triage.
|
| 18 |
+
|
| 19 |
+
A single RoBERTa encoder fine-tuned with LoRA predicts:
|
| 20 |
+
|
| 21 |
+
- Ticket Category
|
| 22 |
+
- Ticket Priority
|
| 23 |
+
- Customer Sentiment
|
| 24 |
+
|
| 25 |
+
All predictions are produced in a single forward pass.
|
| 26 |
+
|
| 27 |
+
---
|
| 28 |
+
|
| 29 |
+
## Model Overview
|
| 30 |
+
|
| 31 |
+
| Task | Example Labels |
|
| 32 |
+
|--------|--------|
|
| 33 |
+
| Category | account, billing, technical, shipping |
|
| 34 |
+
| Priority | low, medium, high, critical |
|
| 35 |
+
| Sentiment | positive, neutral, negative |
|
| 36 |
+
|
| 37 |
+
Architecture:
|
| 38 |
+
|
| 39 |
+
```text
|
| 40 |
+
RoBERTa Base
|
| 41 |
+
│
|
| 42 |
+
▼
|
| 43 |
+
Shared Encoder
|
| 44 |
+
│
|
| 45 |
+
┌────┼────┐
|
| 46 |
+
▼ ▼ ▼
|
| 47 |
+
Category
|
| 48 |
+
Priority
|
| 49 |
+
Sentiment
|
| 50 |
+
Heads
|