CoolFace
Modelpublic

lemonsterpie/finetuned-model-movie-review-sentiment-analysis

sourceHugging Faceupdated 1y agoView on Hugging Face
1likes8downloads
Model Card

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

finetuned-model-movie-review-sentiment-analysis

This model achieves the following results on the evaluation set:

  • —Loss: 0.2338
  • —Accuracy: 0.95
  • —F1: [0.946236559139785, 0.9532710280373832]
  • —Precision: [0.946236559139785, 0.9532710280373832]
  • —Recall: [0.946236559139785, 0.9532710280373832]
  • —Auc: 0.9885

Model description

Binary text classification model that is fine tuned and optimized from JamesH/Moviereviewsentimentanalysismodel.

Intended uses & limitations

Movie review sentiment analysis

Training and evaluation data

Trained on train dataset from datasets library IMDB datasets

Training procedure

200 datapoints from preprocessed train dataset were selected for training.

Training hyperparameters

The following hyperparameters were used during training:

  • —learning_rate: 1.2090707659718472e-05
  • —trainbatchsize: 4
  • —evalbatchsize: 4
  • —seed: 42
  • —gradientaccumulationsteps: 4
  • —totaltrainbatch_size: 16
  • —optimizer: Use OptimizerNames.ADAMW8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
  • —lrschedulertype: linear
  • —lrschedulerwarmup_ratio: 0.09356655990374085
  • —num_epochs: 4
  • —mixedprecisiontraining: Native AMP

Training results

Training LossEpochStepValidation LossAccuracyF1PrecisionRecallAuc
No log1.0130.29630.935[0.9281767955801105, 0.9406392694063926][0.9545454545454546, 0.9196428571428571][0.9032258064516129, 0.9626168224299065]0.9884
No log2.0260.23380.95[0.946236559139785, 0.9532710280373832][0.946236559139785, 0.9532710280373832][0.946236559139785, 0.9532710280373832]0.9885
No log3.0390.26410.94[0.9347826086956522, 0.9444444444444444][0.945054945054945, 0.9357798165137615][0.9247311827956989, 0.9532710280373832]0.9878
No log4.0520.26270.94[0.9347826086956522, 0.9444444444444444][0.945054945054945, 0.9357798165137615][0.9247311827956989, 0.9532710280373832]0.9878

Framework versions

  • —Transformers 4.53.1
  • —Pytorch 2.7.1+cu118
  • —Datasets 3.6.0
  • —Tokenizers 0.21.2