Skip to main content
tez4
Projects
IMDB
Reports
Classifying Reviews using Word-Embeddings

Classifying Reviews using Word-Embeddings

IMDb Reviews are classified (positive, negative) using word-embeddings that are both pretrained and self-made. Neural network architectures include CNN and RNN.
Joël Grosjean
Created on January 7|Last edited on January 29

02468101214epoch0.40.50.60.70.80.9
worthy-sound-21 val_accuracy
worthy-sound-21 accuracy
astral-silence-20 val_accuracy
astral-silence-20 accuracy
driven-shape-19 val_accuracy
driven-shape-19 accuracy
graceful-hill-18 val_accuracy
graceful-hill-18 accuracy
visionary-fog-17 val_accuracy
visionary-fog-17 accuracy
fanciful-haze-16 val_accuracy
fanciful-haze-16 accuracy
firm-vortex-15 val_accuracy
firm-vortex-15 accuracy
sage-sponge-13 val_accuracy
sage-sponge-13 accuracy
wise-capybara-12 val_accuracy
wise-capybara-12 accuracy
spring-glitter-8 val_accuracy
spring-glitter-8 accuracy





Created with ❤️ on Weights & Biases.

https://wandb.ai/tez4/IMDB/reports/Classifying-Reviews-using-Word-Embeddings--VmlldzozMjg1MTkx
Made with Weights & Biases. Sign up or log in to create reports like this one.