final-project-shipdetectionfromsattimage-ccny

Project Structure

• eda : Exploratory data analysis
• meeting_minutes : Meeting for the project.
• model: Code for the project
  • data : data preprocess code
  • evaluation : evaluation code
  • modelweights : modelweights
  • modeling: BasicCNN and FastRCNN codes
  • plots: Plots on the model
• research notes: research notes on the project.

model/data folder

• image_generator.py generates augmeneted images using data_generate.py
• dataset_mapper.py extracts bounding box and image information from the annotations and image files.

model/modeling folder

• BasicCNN.py and BasicCNNPredictor uses dataset_mapper to get the training data. then uses CNN model to train the data and predict the data.
• FastRCNN and FastRCNNtest uses dataset_mapper to get the training data. then uses Fast R CNN model to train the data.

Run the code

pip install -r requrements.txt

python model/modeling/BasicCNN.py