Magnanimous| Spot That Fire V2.0
Team Updates
Firefighters collaborating with FIRE TAMERS
A
Alexandra CáceresCode of kmeans model for virms fire data!
Given a latitude and a longitude, it returns a map of fire spot clusters along the area.
Useful to determine the areas in which there are ongoing fire-related developments.
https://github.com/jmaidana18/spaceapp
J
Jorge Maidana-Montiel This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode characters
| importpandasaspd | |
| fromsklearn.model_selectionimporttrain_test_split | |
| fromsklearn.clusterimportKMeans | |
| importnumpyasnp | |
| importmatplotlib.pyplotasplt | |
| # Lectura del dataset en crudo | |
| data=pd.read_csv("fire_archive_V1_81287.csv") | |
| # Es posible hacer analisis exploratorio en este punto. | |
| # Data Wrangling. | |
| # Discrimina focos de fuego por parametro de Confianza, alta y nominal. | |
| is_high=data['confidence'] =='h' | |
| data_high=data[is_high] | |
| is_nom=data['confidence'] =='n' | |
| data_nom=data[is_nom] | |
| data_proc= [data_high, data_nom] | |
| result=pd.concat(data_proc) | |
| # Visualización de Datos Crudos: Focos por Confidencia. | |
| N=len(result) | |
| colors=np.random.rand(N) | |
| area= (30*np.random.rand(N))**2 | |
| # plt.scatter(data_proc['latitude'], data_proc['longitude'],s=area, c=colors, alpha=0.5) | |
| plt.scatter(result['longitude'],result['latitude']) | |
| plt.title('Paraguay, Focos de Calor, Satelite VIRS') | |
| plt.xlabel('Longitud') | |
| plt.ylabel('Latitud') | |
| plt.show() | |
| plt.savefig('focos_map.png') | |
| # Visualizar datos con focos procesados. | |
| # Division de Dataset en partes. | |
| latTrain, latTest, longTrain, longTest=train_test_split(result['latitude'], result['longitude'], test_size=0.30, random_state=0) | |
| # 30% del dataset es usado para el entrenamiento. | |
| # Entrenamiento de machine learning | |
| train_x=np.zeros((len(latTrain), 2)) | |
| train_x[:,0] =latTrain | |
| train_x[:,1] =longTrain | |
| kmeans=KMeans(n_clusters=100, random_state=0).fit(train_x) | |
| # Prediccion de Zonas de R | |
| y_kmeans=kmeans.predict(train_x) | |
| # Tests de Rendimiento del Algoritmo. | |
| plt.title('Paraguay, Focos de Calor, Satelite VIRS') | |
| plt.xlabel('Longitud') | |
| plt.ylabel('Latitud') | |
| plt.scatter(train_x[:, 1],train_x[:, 0], c=y_kmeans, s=50, cmap='viridis') | |
| plt.savefig('heat_map.png') | |
| plt.title('Paraguay, Focos de Calor, Satelite VIRS') | |
| plt.xlabel('Longitud') | |
| plt.ylabel('Latitud') | |
| plt.scatter(train_x[:, 1], train_x[:, 0], c=y_kmeans, s=50, cmap='viridis') | |
| centers=kmeans.cluster_centers_ | |
| plt.scatter(centers[:, 1], centers[:, 0], c='black', s=200, alpha=0.5); | |
| plt.savefig('heat_map_clusters.png') |
J
Jorge Maidana-MontielG
Guillermo Silva
How to-para usuarios que deseen ejecutar el algoritmo de kmeans Paso 2
J
Jorge Maidana-Montiel
How to-para usuarios que deseen ejecutar el algoritmo de kmeans Paso 1
J
Jorge Maidana-MontielIs hot out there!
J
Jorge Maidana-Montiel