This project is the part of my machine learning internship at Unified Mentor. The goal of this project was to develop a system that can predict the type of forest cover using analysis data for a 30m x 30m patch of land in the forest.
The orginal dataset is an analysis dataset from the forest department performed in the Roosevelt National Forest of northern Colorado.
● 1 - Spruce/Fir
● 2 - Lodgepole Pine
● 3 - Ponderosa Pine
● 4 - Cottonwood/Willow
● 5 - Aspen
● 6 - Douglas-fir
● 7 - Krummholz
● Elevation - Elevation in meters
● Aspect - Aspect in degrees azimuth
● Slope - Slope in degrees
● Horizontal_Distance_To_Hydrology - Horz Dist to nearest surface water features
● Vertical_Distance_To_Hydrology - Vert Dist to nearest surface water features
● Horizontal_Distance_To_Roadways - Horz Dist to nearest roadway
● Hillshade_9am (0 to 255 index) - Hillshade index at 9am, summer solstice
● Hillshade_Noon (0 to 255 index) - Hillshade index at noon, summer solstice
● Hillshade_3pm (0 to 255 index) - Hillshade index at 3pm, summer solstice
● Horizontal_Distance_To_Fire_Points - Horz Dist to nearest wildfire ignition points
● Wilderness_Area (4 binary columns, 0 = absence or 1 = presence) - Wilderness area designation
● Soil_Type (40 binary columns, 0 = absence or 1 = presence) - Soil Type designation
● Cover_Type - Forest Cover Type designation
To achieve accurate classification, I explored various classifying techniques. I finalized Random Forest Classifier for the purpose of this project. The highest achieved average accuracy is 0.88. The model is saved as a pickle file.
Run the Python notebook to train the classifier and save the model as a pickle file.
- Load the .pkl model file and update the path to the model in the
app.pyfile. - Run the
app.pyfile using the command python app.py. - Input the features and run the prediction system. Output label will be displayed.
