def __init__(self,num_filters):
self.featExtractor = Classifiers.IMU_CNN_3D_FEATURE_EXTRACTOR(suffix="40Hz",num_filters=num_filters,patience=250,layers=3,kern_size=32,divide_kernel_size=True)
self.featExtractor.loadBestWeights()
#Output of grid search:
#{'bootstrap': False, 'max_depth': 142, 'max_features': 'auto', 'min_samples_leaf': 2, 'min_samples_split': 2, 'n_estimators': 60}
#Initializing RF
self.clf_RF = RandomForestClassifier(n_estimators=60,min_samples_split=2,min_samples_leaf=2,max_features='auto',max_depth=142,bootstrap=False)
tr_uuids, num_threads)
print("Loading Validation set")
(X_vld, y_vld) = UCI_HAPT.get_all_data_multi_thread_resampling_3D(
vl_uuids, num_threads)
print("Loading test set")
(X_test, y_test) = UCI_HAPT.get_all_data_multi_thread_resampling_3D(
test_uuids, num_threads)
classes = [
"WALKING", "W. UPSTAIRS", "W. DOWNSTAIRS", "SITTING", "STANDING",
"LAYING", "TRANSITION"
]
clf = Classifiers.IMU_CNN_3D_FEATURE_EXTRACTOR(patience=500,
num_filters=nfilters,
layers=n_layers,
kern_size=k,
divide_kernel_size=True,
suffix="40Hz_")
y_train_inv = [np.argmax(y) for y in y_train]
class_weights = class_weight.compute_class_weight('balanced',
np.unique(y_train_inv),
y_train_inv)
d_class_weights = dict(enumerate(class_weights))
clf.fit_class_weights(X_train,
y_train,
X_vld,
y_vld,
batch_size=32,
class_weight=d_class_weights,
epochs=6000)
