README.md

metadata

language: en
tags:
  - banking
  - intent
  - multiclass
datasets:
  - banking77
widget:
  - text: How long until my transfer goes through?
model-index:
  - name: mrm8488/distilroberta-finetuned-banking77
    results:
      - task:
          type: text-classification
          name: Text Classification
        dataset:
          name: banking77
          type: banking77
          config: default
          split: test
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8961038961038961
            verified: true
          - name: Precision Macro
            type: precision
            value: 0.9063619413451185
            verified: true
          - name: Precision Micro
            type: precision
            value: 0.8961038961038961
            verified: true
          - name: Precision Weighted
            type: precision
            value: 0.906361941345118
            verified: true
          - name: Recall Macro
            type: recall
            value: 0.8961038961038963
            verified: true
          - name: Recall Micro
            type: recall
            value: 0.8961038961038961
            verified: true
          - name: Recall Weighted
            type: recall
            value: 0.8961038961038961
            verified: true
          - name: F1 Macro
            type: f1
            value: 0.8914970725184695
            verified: true
          - name: F1 Micro
            type: f1
            value: 0.8961038961038962
            verified: true
          - name: F1 Weighted
            type: f1
            value: 0.8914970725184699
            verified: true
          - name: loss
            type: loss
            value: 0.5607758164405823
            verified: true

distilroberta-base fine-tuned on banking77 dataset for intent classification

Test set accuray: 0.896

How to use

from transformers import AutoTokenizer, AutoModelForSequenceClassification, pipeline

ckpt = 'mrm8488/distilroberta-finetuned-banking77'
tokenizer = AutoTokenizer.from_pretrained(ckpt)
model = AutoModelForSequenceClassification.from_pretrained(ckpt)

classifier = pipeline('text-classification', tokenizer=tokenizer, model=model)
classifier('What is the base of the exchange rates?')
# Output: [{'label': 'exchange_rate', 'score': 0.8509947657585144}]

Created by Manuel Romero/@mrm8488 | LinkedIn Made with ♥ in Spain