roberta-base-squad2 /
Branden Chan
Update to v2
  - squad_v2

roberta-base for QA

NOTE: This is version 2 of the model. See this github issue from the FARM repository for an explanation of why we updated.


Language model: roberta-base
Language: English
Downstream-task: Extractive QA
Training data: SQuAD 2.0
Eval data: SQuAD 2.0
Code: See example in FARM
Infrastructure: 4x Tesla v100


batch_size = 96
n_epochs = 2
base_LM_model = "roberta-base"
max_seq_len = 386
learning_rate = 3e-5
lr_schedule = LinearWarmup
warmup_proportion = 0.2


Evaluated on the SQuAD 2.0 dev set with the official eval script.

"exact": 79.97136359807968
"f1": 83.00449234495325

"total": 11873
"HasAns_exact": 78.03643724696356
"HasAns_f1": 84.11139298441825
"HasAns_total": 5928
"NoAns_exact": 81.90075693860386
"NoAns_f1": 81.90075693860386
"NoAns_total": 5945


In Transformers

from transformers.pipelines import pipeline
from transformers.modeling_auto import AutoModelForQuestionAnswering
from transformers.tokenization_auto import AutoTokenizer

model_name = "deepset/roberta-base-squad2-v2"

# a) Get predictions
nlp = pipeline('question-answering', model=model_name, tokenizer=model_name)
QA_input = {
    'question': 'Why is model conversion important?',
    'context': 'The option to convert models between FARM and transformers gives freedom to the user and let people easily switch between frameworks.'
res = nlp(QA_input)

# b) Load model & tokenizer
model = AutoModelForQuestionAnswering.from_pretrained(model_name)
tokenizer = AutoTokenizer.from_pretrained(model_name)


from farm.modeling.adaptive_model import AdaptiveModel
from farm.modeling.tokenization import Tokenizer
from farm.infer import Inferencer

model_name = "deepset/roberta-base-squad2-v2"

# a) Get predictions
nlp = Inferencer.load(model_name, task_type="question_answering")
QA_input = [{"questions": ["Why is model conversion important?"],
             "text": "The option to convert models between FARM and transformers gives freedom to the user and let people easily switch between frameworks."}]
res = nlp.inference_from_dicts(dicts=QA_input, rest_api_schema=True)

# b) Load model & tokenizer
model = AdaptiveModel.convert_from_transformers(model_name, device="cpu", task_type="question_answering")
tokenizer = Tokenizer.load(model_name)

In haystack

For doing QA at scale (i.e. many docs instead of single paragraph), you can load the model also in haystack:

reader = FARMReader(model_name_or_path="deepset/roberta-base-squad2")
# or 
reader = TransformersReader(model="deepset/roberta-base-squad2",tokenizer="deepset/roberta-base-squad2")


Branden Chan: branden.chan [at] Timo M枚ller: timo.moeller [at] Malte Pietsch: malte.pietsch [at] Tanay Soni: tanay.soni [at]

About us

deepset logo

We bring NLP to the industry via open source!
Our focus: Industry specific language models & large scale QA systems.

Some of our work:

Get in touch: Twitter | LinkedIn | Website