Model description

This is a logistic regression model trained with GPT-2 embeddings on imdb dataset. The notebook to generate this model is in this repository and in this kaggle link.

Intended uses & limitations

This model is trained for educational purposes.

Training Procedure

Hyperparameters

The model is trained with below hyperparameters.

Click to expand

Hyperparameter	Value
memory
steps	[('embedding', HFTransformersLanguage(model_name_or_path='facebook/bart-base')), ('model', LogisticRegression())]
verbose	False
embedding	HFTransformersLanguage(model_name_or_path='facebook/bart-base')
model	LogisticRegression()
embedding__model_name_or_path	facebook/bart-base
model__C	1.0
model__class_weight
model__dual	False
model__fit_intercept	True
model__intercept_scaling	1
model__l1_ratio
model__max_iter	100
model__multi_class	auto
model__n_jobs
model__penalty	l2
model__random_state
model__solver	lbfgs
model__tol	0.0001
model__verbose	0
model__warm_start	False

Model Plot

The model plot is below.

Pipeline(steps=[('embedding',HFTransformersLanguage(model_name_or_path='facebook/bart-base')),('model', LogisticRegression())])

Please rerun this cell to show the HTML repr or trust the notebook.

Evaluation Results

You can find the details about evaluation process and the evaluation results.

Metric	Value
f1_score	0.867535

How to Get Started with the Model

Use the code below to get started with the model.

Click to expand

[More Information Needed]



# Additional Content

## Confusion matrix

![Confusion matrix](confusion_matrix.png)