def do_gridSearch(training_shp, training_image, model_out_path, bands):
PYEO_model.train_cairan_model(image_path=training_image,
shp_path=training_shp,
outModel_path=model_out_path,
bands=bands,
attribute='Id',
shape_projection_id=32630)
def do_grid_search(
training_shape,
training_image,
image_out_path,
model_out_path,
):
PYEO_model.train_cairan_model(image_dir=s0_dir,
outModel_path=outModel,
bands=8,
attribute='Code')
features, classes = get_training_data(training_image,
training_shape,
attribute="Id")
print(features.shape)
print(classes.shape)
X_train, X_test, y_train, y_test = train_test_split(
features.astype(np.uint8),
classes.astype(np.uint8),
train_size=0.7,
test_size=0.3)
# grid search for the best parameters for SVM
# https://scikit-learn.org/stable/auto_examples/svm/plot_rbf_parameters.html
# C_range = np.logspace(0, 1, 2, base=10) # base = 2 for a fine tuning
# gamma_range = np.logspace(0, 128, 2, base=10)
# param_grid = dict(gamma=gamma_range, C=C_range)
# cv = StratifiedShuffleSplit(n_splits=5, test_size=0.2, random_state=42)
# grid = GridSearchCV(SVC(), param_grid=param_grid, cv=cv, n_jobs=6)
# grid.fit(X_train, y_train)
#
# print("The best parameters are %s with a score of %0.2f"
# % (grid.best_params_, grid.best_score_))
# c, gamma = grid.best_params_
# c=
# gamma =
model = svm.SVC(kernel='rbf')
# model = TPOTClassifier(generations=10, population_size=20, verbosity=2,n_jobs=-1)
model.fit(features, classes)
print(model.score(X_test, y_test))
model.export(model_out_path)