metadata
language:
- en
base_model:
- distilbert/distilbert-base-uncased
datasets:
- dair-ai/emotion
metrics:
- accuracy
- f1
library_name: transformers
pipeline_tag: text-classification
tags:
- distilbert
- emotion-classification
- twitter
- text-classification
- pytorch
widget:
- text: I feel so happy and excited today!
- text: I am terrified that something bad will happen.
- text: I really miss the people I love.
model-index:
- name: DistilBERT Twitter Emotion Classifier
results:
- task:
type: text-classification
name: Emotion Classification
dataset:
type: dair-ai/emotion
name: Emotion
split: test
metrics:
- type: accuracy
value: 0.929
name: Test Accuracy
- type: f1
value: 0.929073
name: Test Weighted F1
DistilBERT Twitter Emotion Classifier
This model is a fine-tuned version of
distilbert-base-uncased
for six-class English Twitter emotion classification.