Skip to content

Classification model for CIFAR-10 photo dataset in Keras. Hyperparameter tuning was achieved with the HParams Dashboard and the model was optimized for [1] Number of filters in each CNN layer (HP_NUM_FILTERS1, HP_NUM_FILTERS2, HP_NUM_FILTERS3) [2] Batch size (HP_BAT) [3] Batch Normalization (HP_BATNORM_MOMENT) [4] L2 regularization (HP_L2) [5] E…

License

Notifications You must be signed in to change notification settings

akhtarmf/machinelearning

Repository files navigation

machinelearning

Classification model for CIFAR-10 photo dataset in Keras. Hyperparameter tuning was achieved with the HParams Dashboard and the model was optimized for [1] Number of filters in each CNN layer (HP_NUM_FILTERS1, HP_NUM_FILTERS2, HP_NUM_FILTERS3) [2] Batch size (HP_BAT) [3] Batch Normalization (HP_BATNORM_MOMENT) [4] L2 regularization (HP_L2) [5] Epochs (HP_EPOC) [6] Dropout (HP_DROPOUT) [7] Optimization algorithm (HP_OPTIMIZER)). Refer to Google Colab file, 'ChE788_A4_Q3.ipynb'. Optimized values can be visualized on TensorBoard.

About

Classification model for CIFAR-10 photo dataset in Keras. Hyperparameter tuning was achieved with the HParams Dashboard and the model was optimized for [1] Number of filters in each CNN layer (HP_NUM_FILTERS1, HP_NUM_FILTERS2, HP_NUM_FILTERS3) [2] Batch size (HP_BAT) [3] Batch Normalization (HP_BATNORM_MOMENT) [4] L2 regularization (HP_L2) [5] E…

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published