Edit model card

KEYWORD QUERY VS STATEMENT/QUESTION CLASSIFIER FOR NEURAL SEARCH

Train Loss Validation Acc. Test Acc.
0.000806 0.99 0.997
from transformers import AutoTokenizer, AutoModelForSequenceClassification
  
tokenizer = AutoTokenizer.from_pretrained("shahrukhx01/bert-mini-finetune-question-detection")

model = AutoModelForSequenceClassification.from_pretrained("shahrukhx01/bert-mini-finetune-question-detection")

Trained to add feature for classifying queries between Keyword Query or Question + Statement Query using classification in Haystack

Problem Statement: One common challenge that we saw in deployments: We need to distinguish between real questions and keyword queries that come in. We only want to route questions to the Reader branch in order to maximize the accuracy of results and minimize computation efforts/costs.

Baseline: https://www.kaggle.com/shahrukhkhan/question-v-statement-detection

Dataset: https://www.kaggle.com/stefanondisponibile/quora-question-keyword-pairs

Kaggle Notebook: https://www.kaggle.com/shahrukhkhan/question-vs-statement-classification-mini-bert/

Downloads last month
356,655
Safetensors
Model size
11.2M params
Tensor type
I64
ยท
F32
ยท
Inference API

Spaces using shahrukhx01/bert-mini-finetune-question-detection 2