File size: 1,425 Bytes
de86483
 
 
 
 
 
5aad6cd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
---
license: mit
language:
- en
library_name: pytorch
---
# Athena- Intent

Classifies intent of the query for Athena

## Architecture
distilbert-base-uncased backbone, finetuned over a multiclass classification problem

### Description

Classifies user intent of queries into the following classes:
0: Keyword Search
1: Semantic Search
2: Direct Question Answering


## Uses

This model is intended to be used in Athena for performing QA on enterprise document stores.


## Bias, Risks, and Limitations

Dataset was generated using ChatGPT (gpt-3.5-turbo). It consists of 5000 English sentences and the nature of their intent, annotated manually.

## Usage

```
from transformers import AutoTokenizer
from transformers import TFDistilBertForSequenceClassification
import tensorflow as tf

model = TFDistilBertForSequenceClassification.from_pretrained("sourcerersupreme/athena-intent")
tokenizer = AutoTokenizer.from_pretrained("sourcerersupreme/athena-intent")

class_semantic_mapping = {
        0: "Keyword",
        1: "Semantic",
        2: "QA"
    }

# Get user input
user_query = "What is a CDP?"

# Encode the user input
inputs = tokenizer(user_query, return_tensors="tf", truncation=True, padding=True)

# Get model predictions
predictions = model(inputs)[0]

# Get predicted class
predicted_class = tf.math.argmax(predictions, axis=-1)

print(f"Predicted class: {class_semantic_mapping[int(predicted_class)]}")
```