Proscia Tech challenge


Minimum Software Requirements

  • Python 3

  • OpenCV-3.1.0

  • sklearn-0.19.1

  • numpy-1.14.0

  • matplotlib-2.0.2

  • OS-Windows 10

Files submitted


1.pathology_A : Contains all pathology a type images

2.pathology_B : Contains all pathology b type images

3.tech_challenge:-->identify.py : The main file used for classification task

4. writeup: The short report of observations

5.pathology_A_HSV_analysis: HSV figures of pathology A generated during the exploratory process

6.pathology_B_HSVanalysis: HSV figures of pathology B generated during he exploratory process

Files generated


1.svm_model_poly_trained.sav : The training model generated during training if selected

Installation and Execution


To install and run the program:

1. Unzip the contents
2. cd to proscia/tech_challenge
3. Run python idenify_slide.py [followed by the path of the image you want to test] using command line for example : python identify_slide.py ../pathology_A/10.png
4. The command prompt will ask user to either enter 1 for classifying using ML or enter 0 for classifying using computer vision methods5. Output will be either A or B