---
license: apache-2.0
datasets:
- squad_v2
model-index:
- name: nlpconnect/deberta-v3-xsmall-squad2
  results:
  - task:
      type: question-answering
      name: Question Answering
    dataset:
      name: squad_v2
      type: squad_v2
      config: squad_v2
      split: validation
    metrics:
    - type: exact_match
      value: 79.3917
      name: Exact Match
      verified: true
      verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZTFiMWI5YzFlMDZhMzc2NDIwYjNiZmIyMThmOWQxYjFjZmM2ZDQ0OGM2NmNlNmI3Y2U2N2JjMmVkZTgyZjNiOCIsInZlcnNpb24iOjF9.MCw9UJ3MI3Lf5hvOgk7Lw2xZfN4678p7ebG3vnGXX_Avw6fELTPwxZ9qGA-9tL00p4NxaSb3Cx6XAFvWetAIBA
    - type: f1
      value: 82.6738
      name: F1
      verified: true
      verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiMjdiYWY2MzU4YjZhMWQzZGJhZTk3NzU3Y2UwYmQ4MzliZmQxOGUxZDllN2Y0ZmZhYjVlNTE0MzY1MjU5OWMwMCIsInZlcnNpb24iOjF9.zeWLwXy77n0YKxGA5gjySe8p-_nPQxbiPnvQU2tF45IyMmlYKUuLeq4hJnNe-5NgriTf8xkBJBE7Cr5lWHy_Cw
  - task:
      type: question-answering
      name: Question Answering
    dataset:
      name: squad
      type: squad
      config: plain_text
      split: validation
    metrics:
    - type: exact_match
      value: 84.9246
      name: Exact Match
      verified: true
      verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZGJhYmU0Y2I4Y2UyOGVlOTlkMmQ2OTcyMTZkNTkwNTMzNzhmNzZiYjU4ZDkxMGM5NzAyMjk1M2ExNGIzOWU4NCIsInZlcnNpb24iOjF9.ql1rCId6lQ7Uwq2spG3q2fFppkFGHA1IWQjvyPRhvKdRNzApBO0mu9JjMAv4uNKZX-kmGEkI018_9tAzN7kwDw
    - type: f1
      value: 91.6201
      name: F1
      verified: true
      verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZjBjMmI0OTFmODVjMzllZDM0NTdmNjU4NGI4NzA4NTJhOWVkMDQ5OTY0MDcyMWEwZTFkODNlY2VhZjU2NWJmZSIsInZlcnNpb24iOjF9.rGvF60bfWIXzB66C7fkdxCtZvRZ_m3onbLaNbs7M4M0Fk27xnMat6IAy1DeTztkOKLoiD2s2NQH6wXid83cgCw
---

# Deberta-v3-xsmall-squad2


## What is SQuAD?

Stanford Question Answering Dataset (SQuAD) is a reading comprehension dataset, consisting of questions posed by crowdworkers on a set of Wikipedia articles, where the answer to every question is a segment of text, or span, from the corresponding reading passage, or the question might be unanswerable.

SQuAD2.0 combines the 100,000 questions in SQuAD1.1 with over 50,000 unanswerable questions written adversarially by crowdworkers to look similar to answerable ones. To do well on SQuAD2.0, systems must not only answer questions when possible, but also determine when no answer is supported by the paragraph and abstain from answering.

## Inference

```python

from transformers import pipeline

qa = pipeline("question-answering", model="nlpconnect/deberta-v3-xsmall-squad2")

result = qa(context="My name is Sarah and I live in London", question="Where do I live?")
```

## Accuracy

```json
squad_v2 = {'exact': 79.392,
   'f1': 82.674}
   
squad = {'exact': 84.925,
   'f1': 91.620}
 ```