Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- text-classification
|
| 5 |
+
- distilbert
|
| 6 |
+
- conversation-classification
|
| 7 |
+
- knowledge-distillation
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
# ThradBertConversationClassifier
|
| 11 |
+
|
| 12 |
+
Personal copy of a BERT model for conversation intent classification I developed at Thrad.ai
|
| 13 |
+
|
| 14 |
+
## Model Details
|
| 15 |
+
|
| 16 |
+
- **Base Architecture**: DistilBERT
|
| 17 |
+
- **Task**: Multi-class conversation intent classification
|
| 18 |
+
|
| 19 |
+
## Usage
|
| 20 |
+
|
| 21 |
+
```python
|
| 22 |
+
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
| 23 |
+
|
| 24 |
+
model = AutoModelForSequenceClassification.from_pretrained("ScottBiggs2/ThradBertConversationClassifier")
|
| 25 |
+
tokenizer = AutoTokenizer.from_pretrained("ScottBiggs2/ThradBertConversationClassifier")
|
| 26 |
+
|
| 27 |
+
# Example inference
|
| 28 |
+
text = "Your conversation text here"
|
| 29 |
+
inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=512)
|
| 30 |
+
outputs = model(**inputs)
|
| 31 |
+
predictions = outputs.logits.softmax(dim=-1)
|
| 32 |
+
```
|
| 33 |
+
|