Jean-Baptiste

/

roberta-large-ner-english

like 68

Hi guys,

This model is amazing. Congratulations and thanks for sharing \o/

I have deployed the model with the option "Use in Transformers" using the code below and I got the same output as presented in model card:

from transformers import AutoTokenizer, AutoModelForTokenClassification
tokenizer = AutoTokenizer.from_pretrained("Jean-Baptiste/roberta-large-ner-english")
model = AutoModelForTokenClassification.from_pretrained("Jean-Baptiste/roberta-large-ner-english")

##### Process text sample (from wikipedia)

from transformers import pipeline
nlp = pipeline('ner', model=model, tokenizer=tokenizer, aggregation_strategy="simple")
nlp("Apple was founded in 1976 by Steve Jobs, Steve Wozniak and Ronald Wayne to develop and sell Wozniak's Apple I personal computer")

[{'entity_group': 'ORG',
  'score': 0.99381506,
  'word': ' Apple',
  'start': 0,
  'end': 5},
 {'entity_group': 'PER',
  'score': 0.99970853,
  'word': ' Steve Jobs',
  'start': 29,
  'end': 39},
 {'entity_group': 'PER',
  'score': 0.99981767,
  'word': ' Steve Wozniak',
  'start': 41,
  'end': 54},
 {'entity_group': 'PER',
  'score': 0.99956465,
  'word': ' Ronald Wayne',
  'start': 59,
  'end': 71},
 {'entity_group': 'PER',
  'score': 0.9997918,
  'word': ' Wozniak',
  'start': 92,
  'end': 99},
 {'entity_group': 'MISC',
  'score': 0.99956393,
  'word': ' Apple I',
  'start': 102,
  'end': 109}]

However, When deploying with the Amazon SageMaker option I got a different output as shown bellow:

from sagemaker.huggingface import HuggingFaceModel
import sagemaker

role = sagemaker.get_execution_role()
# Hub Model configuration. https://huggingface.co/models
hub = {
    'HF_MODEL_ID':'Jean-Baptiste/roberta-large-ner-english',
    'HF_TASK':'token-classification'
}

# create Hugging Face Model Class
huggingface_model = HuggingFaceModel(
    transformers_version='4.17.0',
    pytorch_version='1.10.2',
    py_version='py38',
    env=hub,
    role=role, 
)

# deploy model to SageMaker Inference
predictor = huggingface_model.deploy(
    initial_instance_count=1, # number of instances
    instance_type='ml.m5.xlarge' # ec2 instance type
)

predictor.predict({
    'inputs': "Apple was founded in 1976 by Steve Jobs, Steve Wozniak and Ronald Wayne to develop and sell Wozniak's Apple I personal computer"
})

[{'entity': 'ORG',
  'score': 0.9938150644302368,
  'index': 1,
  'word': 'ĠApple',
  'start': 0,
  'end': 5},
 {'entity': 'PER',
  'score': 0.9996582269668579,
  'index': 7,
  'word': 'ĠSteve',
  'start': 29,
  'end': 34},
 {'entity': 'PER',
  'score': 0.999758780002594,
  'index': 8,
  'word': 'ĠJobs',
  'start': 35,
  'end': 39},
 {'entity': 'PER',
  'score': 0.9997087121009827,
  'index': 10,
  'word': 'ĠSteve',
  'start': 41,
  'end': 46},
 {'entity': 'PER',
  'score': 0.9998393058776855,
  'index': 11,
  'word': 'ĠW',
  'start': 47,
  'end': 48},
 {'entity': 'PER',
  'score': 0.9998897314071655,
  'index': 12,
  'word': 'oz',
  'start': 48,
  'end': 50},
 {'entity': 'PER',
  'score': 0.9998410940170288,
  'index': 13,
  'word': 'ni',
  'start': 50,
  'end': 52},
 {'entity': 'PER',
  'score': 0.9998093247413635,
  'index': 14,
  'word': 'ak',
  'start': 52,
  'end': 54},
 {'entity': 'PER',
  'score': 0.9995868802070618,
  'index': 16,
  'word': 'ĠRonald',
  'start': 59,
  'end': 65},
 {'entity': 'PER',
  'score': 0.9995424747467041,
  'index': 17,
  'word': 'ĠWayne',
  'start': 66,
  'end': 71},
 {'entity': 'PER',
  'score': 0.9997656941413879,
  'index': 22,
  'word': 'ĠW',
  'start': 92,
  'end': 93},
 {'entity': 'PER',
  'score': 0.999869704246521,
  'index': 23,
  'word': 'oz',
  'start': 93,
  'end': 95},
 {'entity': 'PER',
  'score': 0.9997791647911072,
  'index': 24,
  'word': 'ni',
  'start': 95,
  'end': 97},
 {'entity': 'PER',
  'score': 0.9997527003288269,
  'index': 25,
  'word': 'ak',
  'start': 97,
  'end': 99},
 {'entity': 'MISC',
  'score': 0.9996131062507629,
  'index': 27,
  'word': 'ĠApple',
  'start': 102,
  'end': 107},
 {'entity': 'MISC',
  'score': 0.9995146989822388,
  'index': 28,
  'word': 'ĠI',
  'start': 108,
  'end': 109}]

What do I need to change on SageMaker to get the same amazing result please?