From 6ede0bc555d2b1a3fe8985580acba300028ad83f Mon Sep 17 00:00:00 2001
From: mlang <marc.lang@teledetection.fr>
Date: Wed, 12 Dec 2018 12:13:10 +0100
Subject: [PATCH] Add of a second recipe
update recipe
---
.gitignore | 4 +
scripts/classification_from_segmentation.py | 127 +++++++++++++++++++-
2 files changed, 130 insertions(+), 1 deletion(-)
diff --git a/.gitignore b/.gitignore
index c6c5a74..b1a12f3 100755
--- a/.gitignore
+++ b/.gitignore
@@ -60,3 +60,7 @@ data/l8_20130707\.tif\.aux\.xml
data/l8_20130831\.tif\.aux\.xml
qgis-auth\.db
+
+\.spyderproject
+
+\.spyderworkspace
diff --git a/scripts/classification_from_segmentation.py b/scripts/classification_from_segmentation.py
index a5b7ea4..ba8d170 100644
--- a/scripts/classification_from_segmentation.py
+++ b/scripts/classification_from_segmentation.py
@@ -20,6 +20,132 @@ except:
print 'pandas not imported'
+
+def my_recipe2(**kwargs):
+
+ stats = kwargs.get('stats', ['mean', 'std'])
+ field_id = kwargs.get('field_id', 'DN')
+ field_class = kwargs.get('field_class', 'class')
+ train_size = kwargs.get('train_size', 0.5)
+ nodata = kwargs.get('nodata', None)
+ target_names = kwargs.get('target_names')
+ debug = kwargs.get('debug', False)
+ field_pred = kwargs.get('field_pred', 'pred_class')
+
+ n_jobs = kwargs.get('n_jobs', 1)
+ save = kwargs.get('save', False)
+ X_train = kwargs.get('X_train')
+ Y_train = kwargs.get('Y_train')
+ X_test = kwargs.get('X_test')
+ Y_test = kwargs.get('Y_test')
+ X = Y = FID = None
+ FID_train = None
+ FID_test = None
+ i=0
+ base = kwargs.get('base')
+ X = np.load(base + '_stats_allsegment_X{}.npy'.format(i))
+ Y = np.load(base + '_stats_allsegment_Y{}.npy'.format(i))
+ FID = np.load(base + '_stats_allsegment_FID{}.npy'.format(i))
+
+ #X_train = np.load(base + '_X_train.npy')
+ #Y_train = np.load(base + '_Y_train.npy')
+ #X_test = np.load(base + '_X_test.npy')
+ #Y_test = np.load(base + '_Y_test.npy')
+
+ dict_mat = {}
+ dict_mat['nb_of_shp'] = 1
+ dict_mat['gml'] = False
+
+ if kwargs.get('cross_validation'):
+ if save :
+ base = kwargs.get('base')
+ base += 'cross_validation.csv'
+ else:
+ base = None
+ X_samp = kwargs.get('X_samp')
+ Y_samp = kwargs.get('Y_samp')
+ FID_samp = kwargs.get('FID_samp')
+ temp_result = train_test_split(X_samp, Y_samp, FID_samp,
+ train_size=train_size)
+ (X_train, X_test, Y_train, Y_test,
+ FID_train, FID_test) = temp_result
+ scaler = StandardScaler().fit(X_train)
+ cross_validation_dict = kwargs.get('cross_validation_dict')
+ clf = cla.select_best_svm_classifier_sample(scaler.transform(X_train),
+ Y_train,
+ output_result_name=base,
+ **cross_validation_dict)
+ estimator = clf.best_estimator_
+ if save:
+ base = kwargs.get('base')
+
+ X_train_name = base + '_X_train.npy'
+ X_test_name = base + '_X_test.npy'
+ Y_train_name = base + '_Y_train.npy'
+ Y_test_name = base + '_Y_test.npy'
+ FID_train_name = base + '_FID_train.npy'
+ FID_test_name = base + '_FID_test.npy'
+
+ np.save(X_train_name, X_train)
+ np.save(X_test_name, X_test)
+ np.save(Y_train_name, Y_train)
+ np.save(Y_test_name, Y_test)
+ np.save(FID_train_name, FID_train)
+ np.save(FID_test_name, FID_test)
+ else:
+ X_train = kwargs.get('X_train')
+ Y_train = kwargs.get('Y_train')
+ X_test = kwargs.get('X_test')
+ Y_test = kwargs.get('Y_test')
+ scaler = StandardScaler().fit(X_train)
+ estimator = kwargs.get('estimator', None)
+
+ dict_mat['X'], dict_mat['Y'], dict_mat['FID'] = scaler.transform(X), Y, FID
+ dict_mat['X_train'], dict_mat['X_test'] = (scaler.transform(X_train),
+ scaler.transform(X_test))
+ dict_mat['Y_train'], dict_mat['Y_test'] = Y_train, Y_test
+
+ temp_result = cla.classifier(scaler.transform(X_train), Y_train,
+ scaler.transform(X_test),
+ Y_test, scaler.transform(X), dict_mat,
+ estimator=estimator, n_jobs=n_jobs)
+ Y_predict, classif, dict_mat, classif_parameters = temp_result
+
+ print 'Control : classifier done'
+ print classif_parameters
+
+ # Get some quality indices
+ cm, report, accuracy = cla.pred_error_metrics(Y_predict, Y_test,
+ target_names=target_names)
+
+ kappa = cohen_kappa_score(Y_predict, Y_test)
+
+ dict_output = {'cm': cm,
+ 'kappa': kappa,
+ 'accuracy': accuracy,
+ 'report': report,
+ 'Y_predict': Y_predict,
+ 'Y_true': Y_test,
+ 'classif': classif,
+ 'FID': FID,
+ 'estimator': str(classif_parameters)}
+
+ # dict_output.update(dict_mat)
+ dict_output = dict(dict_output, **dict_mat)
+
+ if save:
+ base = kwargs.get('base')
+ try:
+ fichier = open(base + '_classifier.txt', 'w')
+ fichier.write(dict_output['estimator'])
+ fichier.close()
+ except:
+ print 'classifier not saved'
+ save_output(dict_output, base + '_classif')
+
+ return dict_output
+
+
def my_recipe(segmentation, roi_file, raster,
**kwargs):
"""
@@ -591,7 +717,6 @@ def classification_from_A_to_Z(segmentation, roi_file, raster,
dict_mat['FID_{}'.format(i+1)],
field_pred=field_pred, field_id=field_id)
print 'Control : vector updated'
-
# Get some quality indices
cm, report, accuracy = cla.pred_error_metrics(Y_predict, Y_test,
target_names=target_names)
--
GitLab