jagui1 / aiproject3 Goto Github PK

Jeremy Aguillon
CMSC 471
Project 3

                             Contents:

pickling.py - program that pickles the given data into data.pickle
            - not generalizible only created based on specific data set

data.pickle - pickled data that contains all of the featurized images and their labels

proj3.ipynb - program that creates the confusion matrix and creates a graph of it for visualization

classify.py - this is the requested program that takes in a command line argument for the path to the image to test and predicts the type of image
            - uses data.pickle instead of folders of files so it is more generalized
	    - can use any other data set in the form of [ <list of vectorized images>, <list of corresponding labels> ] named pickle.data

project3Report.tex - tex source code for the report

project3Report.pdf - readable version of the report


                            To install:

My program uses scikitlearn and all the imports are already at the top of my programs.
If you need to install scikit learn =>

From their website 

Scikit-learn requires:
Python (>= 2.6 or >= 3.3), NumPy (>= 1.6.1), SciPy (>= 0.9)

This is how I did 

conda install scikit-learn

Another option

pip install -U scikit-learn


                            To Run:

python/python3 classify.py <path to img>


                            Examples:

python3 classify.py "../Training/00/01.jpg"

python classify.py "test.jpg"