An ultra-lightweight model useful for distinguishing between different handwritten digits. All in less than 1.8 MiB, and gives a validation accuracy of 98.60% on Kaggle's private testing dataset.
- Downloads last month
- 3
An ultra-lightweight model useful for distinguishing between different handwritten digits. All in less than 1.8 MiB, and gives a validation accuracy of 98.60% on Kaggle's private testing dataset.
# Available backend options are: "jax", "torch", "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" import keras model = keras.saving.load_model("hf://DevBhuyan/Digit-Recognition-Lightweight")