metadata
license: llama3.3
datasets:
- stanfordnlp/imdb
metrics:
- accuracy
base_model:
- distilbert/distilbert-base-uncased
pipeline_tag: text-classification
This model is an extra tuned MLM DistilBert model, used for natural language processing.
It was trained on the full dataset of IMDB for 5 epochs. It is able to classify wheter the text input, a movie review, is positive or negative.
It's current accuracy is 92.7%. In comparison, a DistilBert model trained on a subset of IMDB data and for 3 epochs only, has 87% accuracy.