Update README.md
Browse files
README.md
CHANGED
|
@@ -1,6 +1,36 @@
|
|
| 1 |
---
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
-
|
| 4 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
pipeline_tag: text-classification
|
| 6 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
+
language: en
|
| 3 |
license: apache-2.0
|
| 4 |
+
base_model: distilbert-base-uncased
|
| 5 |
+
tags:
|
| 6 |
+
- text-classification
|
| 7 |
+
- sentiment-analysis
|
| 8 |
+
- transformer
|
| 9 |
+
- distilbert
|
| 10 |
+
- huggingface
|
| 11 |
pipeline_tag: text-classification
|
| 12 |
+
widget:
|
| 13 |
+
- text: "I love the way your startup handled my issue. Brilliant support!"
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
# 💫 Sentiment Model Aura (KS-Vijay)
|
| 17 |
+
|
| 18 |
+
This model is fine-tuned using **DistilBERT** to classify the **sentiment of startup grievance text** as either `Positive`, `Neutral`, or `Negative`.
|
| 19 |
+
|
| 20 |
+
## 💼 Use Case
|
| 21 |
+
|
| 22 |
+
Used in a smart Grievance Redressal System to analyze tone and emotion of user complaints, providing faster triage and better prioritization.
|
| 23 |
+
|
| 24 |
+
## 🔍 Model Summary
|
| 25 |
+
|
| 26 |
+
- **Task:** Sentiment Classification
|
| 27 |
+
- **Model:** DistilBERT
|
| 28 |
+
- **Labels:** `Positive`, `Neutral`, `Negative`
|
| 29 |
+
- **Format:** `safetensors`
|
| 30 |
+
- **Trained on:** Complaints from `complaints.csv` (custom dataset)
|
| 31 |
+
- **Framework:** PyTorch with 🤗 Transformers
|
| 32 |
+
|
| 33 |
+
## 🧪 Example Input
|
| 34 |
+
|
| 35 |
+
```text
|
| 36 |
+
"The support staff was helpful, but the delay was frustrating."
|