DAS_ice

Here I provide a code example to generate features using InceptionV3, load the trained model and classify some scalograms. There are some packages that need to be installed in order to run the code: numpy, tensorflow, scikit-learn, tqdm, yellowbrick, etc. All functions are inside "run_all.py". final_images contain scalograms for the entire two weeks for two channnels: 00350 (distance = 0.59 km) and 15550 (dist = 31.62 km) python run_all.py should : (i) generate a graphic showing that there are two clusters for the scalograms in final_images according with the Calinski Index. (ii) generature features for each scalogram, recognize if scalogram belong to cluster 0 or 1 based on those features and on the trained model (.pkl file). clust_img will contain the scalograms for each cluster and the results are saved to "total_clusts.csv"