Gender Classification by Name Model Details Model Name : Genderize Developed By : Imran Ali Model Type : Text Classification Language : English License : MIT Description This model classifies gender based on the input name. It uses a pre trained BERT model as the base and has been fine tuned on a dataset of names and their associated genders. Training Details Training Data : Dataset of names and genders (e.g., Dannel gender name dataset) Training Procedure : Fine tuned using BERT model with a classification head Training Hyperparameters : Batch size: 8 Gradient accumulation steps: 1 learning rate: 2e 5 Total steps: 20,005 Number of trainable parameters: 109,483,778 (1.9M) Evaluation Testing Data : Split from the training dataset Metrics : Accuracy, Precision, Recall, F1 Score Uses Direct Use : Classifying the gender of a given name Downstream Use : Enhancing applications that require gender identification based on names (e.g., personalized marketing, user profiling) Out of Scope Use : Using the model for purposes other than gender classification without proper validation Bias, Risks, and Limitations Bias : The model may reflect biases present in the training data. It is important t…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy