Skip to content

Latest commit

 

History

History
12 lines (10 loc) · 844 Bytes

File metadata and controls

12 lines (10 loc) · 844 Bytes

Effect of Dropout on Model Performance toward skin cancer classification

We will analyze the effect of using Dropout in the visible and hidden layers while training the models on the skin cancer dataset.

Prerequisites:

Python 3.5 Keras 2.2.0 Tensorflow-GPU 1.9.0 Scikit-Learn

Acknowledgment

Feel free to run the codes from the attached Jupyter Notebook. The codes can be used with your own custom datasets. All that you need is a CSV file containing the features and class labels. We believe the code would be of good value for the research community and request to kindly cite our study: T. R. Thamizhvani, Suganthi Lakshmanan & R. Sivaramakrishnan (2018). Mobile application-based computer-aided diagnosis of skin tumours from dermal images, The Imaging Science Journal, 66:6, 382-391, DOI: 10.1080/13682199.2018.1492682