By using a convolutional neural network (CNN) this program classifies microcalcifications and masses in a mammogram as either benign or malignant. If there are no masses present in the breast tissue then the mammogram will be classified as normal. The biggest challenge in this project was that the amount of mammogram images without any segmentation or overlay available online (publicly) was scarce and not feasible to train an entire CNN on, thus why I trained the last layer of Google's Inception v3 network on pre-segmented image data. The .gitignore contains files that are installed via the retrain.py script that is pulled from TensorFlow.
