def convertSimpleToInput(simple):
inputWords=simple.inputWords
labelsList=[]
inputWordsList=[]
for w in inputWords:
labelsList.append(w.label)
inputWordsList.append(w.text)
return inputData.inputData("1",inputWordsList,labelsList)
def __init__(self, interface):
self.interface = interface
self.table = vtk.vtkTable
self.colorBar = {
'Point1': [0, 0, 1, 0],
'Point2': [0.5, 1, 1, 0],
'Point3': [1, 1, 0, 0]
}
self.input_Data = inputData.inputData()
def get_normals(vtkclassdict):
inputdata = inputData.inputData()
labels = []
dataset_concatenated = []
# This looks really confusing but is really not
for folderclass, vtklist in vtkclassdict.items():
try:
with open(folderclass + ".pickle", 'rb') as f:
dataset = pickle.load(f)
normal_features = []
for vtkfilename in vtklist:
#We'll load the same files and get the normals
features = inputdata.load_features(
vtkfilename, feature_points=["Normals"])
normal_features.append(features)
#This reshaping stuff is to get the list of points, i.e., all connected points
#and the corresponding label which is the normal in this case
#The data in the dataset contains lists with different sizes
normal_features = np.array(normal_features)
featshape = np.shape(normal_features)
labels.extend(
normal_features.reshape(featshape[0], featshape[1], -1))
dsshape = np.shape(dataset)
dataset_concatenated.extend(
dataset.reshape(dsshape[0], dsshape[2], dsshape[3], -1))
except Exception as e:
print('Unable to process', pickle_file, ':', e)
raise
# lens = np.array([len(dataset_concatenated[i]) for i in range(len(dataset_concatenated))])
# mask = np.arange(lens.max()) < lens[:,None]
# padded = np.zeros(mask.shape + (3,))
# padded[mask] = np.vstack((dataset_concatenated[:]))
# return np.array(padded), np.array(labels)
return np.array(dataset_concatenated), np.array(labels)
def main():
#read the config file
config = json.load(open('config.json'))
#initalte the class
try:
dataClass = inputData(config)
except FileNotFoundError as e:
print(e.__str__())
exit()
# Treating the file to adhere to the given formats
dataClass.treatColsDataType()
# for col in dataClass.Dataframe.columns:
# print(col, type(dataClass.Dataframe.loc[0,col]))
# For each collection modify the startup Details
for collections in dataClass.validator['collections']:
collectionName = list(collections.keys())[0]
collectionDetails = list(collections.values())[0]
#Modify database
infoDict = dataClass.modifyInfo(collectionName, collectionDetails)
dataClass.updateDatabase(collectionName, infoDict)
#Validate and filter the non conforming data. Let us not bother too much now
# startupData.validatedData()
#change the structure of startupInfo
# startupInfoList = startupData.modifyStartupInfo()
#
# insert data in collection
# startupData.updateDatabase('startupInfo',startupInfoList)
# change the structure of startupInfo
# fundingInfoList = startupData.modifyFundingInfo()
# insert data in collection
# startupData.updateDatabase('fundingInfo', fundingInfoList)
print("Finished updating the database")
print()
import numpy as np
from vtk.util import numpy_support
parser = argparse.ArgumentParser(description='Shape Variation Analyzer')
parser.add_argument('--dataPath',
action='store',
dest='dirwithSub',
help='folder with subclasses',
required=True)
if __name__ == '__main__':
args = parser.parse_args()
dataPath = args.dirwithSub
inputdata = inputData.inputData()
data_folders = inputdata.get_folder_classes_list(dataPath)
print(data_folders)
polydata = vtk.vtkPolyData()
for datafolders in data_folders:
i = 0
vtklist = glob.glob(os.path.join(datafolders, "*.vtk"))
print(vtklist)
matfile = glob.glob(os.path.join(datafolders, "*.mat"))
matfile_str = ''.join(map(str, matfile))
print('matfile', matfile_str)
for matlabfilename in matfile:
mat_contents = sio.loadmat(matlabfilename, squeeze_me=True)
def main():
print("Cut Calculator by Pyton")
inputData()
import anneal
import inputData
if __name__ == '__main__':
inputData.inputData()
totalDistanceAnneal, pathAnneal = anneal.anneal()
#print("Stimulated Annealing Results - ")
#print("The path to be taken is:\n")
print(" ".join(str(x) for x in pathAnneal))
#print("\nThe total distance covered will be: "+str(totalDistanceAnneal)+" units\n")
def generate(args):
np.set_printoptions(threshold='nan')
print('###########In generation shape#############')
#
#dataPathtrain=args.dirwithSubtrain
#dataPathtest=args.dirwithSubtest
pickle_file = args.picklefile
pickle_file_output = args.pickle_file_new
# Get the data from the folders with vtk files
inputdata = inputData.inputData()
fi = open(pickle_file, 'rb')
dataset = pickle.load(fi)
test_labels = dataset["test_labels"]
train_labels = dataset["train_labels"]
valid_labels = dataset["valid_labels"]
test_dataset = dataset["test_dataset"]
train_dataset = dataset["train_dataset"]
valid_dataset = dataset["valid_dataset"]
print(test_labels)
#data_folders_train = inputdata.get_folder_classes_list(dataPathtrain)
#data_folders_test = inputdata.get_folder_classes_list(dataPathtest)
#pickled_datasets_train,vtklisttrain = inputdata.maybe_pickle(data_folders_train, 6, feature_points=args.feature_names)
#pickled_datasets_test,vtklisttest = inputdata.maybe_pickle(data_folders_test, 0, feature_points=args.feature_names)
#Create the labels, i.e., enumerate the groups
#dataset_train,labels_train = get_labels(pickled_datasets_train)
#print('pickled_datasets_train',pickled_datasets_train,'pickled_datasets_test',pickled_datasets_test)
#dataset_test,labels_test = get_labels(pickled_datasets_test)
# Compute the total number of shapes and train/test size
total_number_shapes_train = train_dataset.shape[0]
total_number_shapes_test = test_dataset.shape[0]
print('total number of shapes train', total_number_shapes_train)
print('total number of shapes test', total_number_shapes_test)
print('labels to train', train_labels, 'labels to test', test_labels)
#num_train = int(args.train_size*total_number_shapes_train)
#num_valid = int((total_number_shapes_train - num_train)*args.validation_size)
# Randomize the original dataset
#print('shape before randomize',dataset_train.shape)
shuffled_dataset, shuffled_labels = inputdata.randomize(
train_dataset, train_labels)
#print('shape after randomize',shuffled_dataset.shape)
#shuffled_dataset_test,shuffled_labels_test = inputdata.randomize(dataset_test,labels_test)
#shuffled_dataset = np.reshape(shuffled_dataset, (total_number_shapes_train, -1))
#print('shape after reshape',shuffled_dataset.shape)
#shuffled_dataset_test = np .reshape(shuffled_dataset_test,(total_number_shapes_test,-1))
# Generate SMOTE with out including the valid/test samples, in some cases, this may raise an error
# as the number of samples in one class is less than 5 and SMOTE cannot continue. Just run it again
dataset_res, labels_res = generate_with_SMOTE(shuffled_dataset,
shuffled_labels)
# SANITY CHECKS
print('dataset train', np.shape(train_dataset))
print('labels train', np.shape(train_labels))
#print('dataset_res',np.shape(dataset_res))
#print('labels_res',np.shape(labels_res))
#print('num_train', num_train)
#print('num_valid', num_valid)
print('number of labels', np.shape(np.unique(train_labels)))
#print('number of labels resampled',np.shape(np.unique(labels_res)))
#print('Labels resampled',np.unique(labels_res).tolist())
print('test labels', test_labels)
#SVM_classification(dataset_res,labels_res,dataset_test,labels_test)
#clf=LinearSVC(random_state=0)
#clf=GaussianProcessClassifier(1.0 * RBF(1.0))
#clf.fit(dataset_res,labels_res)
#prediction = clf.predict(dataset_test)
#for i in range(0,total_number_shapes_test):
# head,tail = os.path.split(vtklisttest[i])
# print(tail,prediction[i])
#PCA_plot(dataset,labels,dataset_res,labels_res)
try:
f = open(pickle_file_output, 'wb')
save = {
#'train_dataset': dataset_res,
#'train_labels': labels_res,
'train_dataset': dataset_res,
'train_labels': labels_res,
'valid_dataset': valid_dataset,
'valid_labels': valid_labels,
# 'test_dataset': dataset[num_train + num_valid:],
#'test_labels': labels[num_train + num_valid:]
'test_dataset': test_dataset,
'test_labels': test_labels
}
pickle.dump(save, f, pickle.HIGHEST_PROTOCOL)
#f.close()
except Exception as e:
print('Unable to save data to', pickle_file, ':', e)
raise
# -*- coding: utf-8 -*- """ Created on Tue May 16 00:40:26 2017 @author: Administrator """ from multiprocessing.dummy import Pool as ThreadPool import numpy as np import pandas as pd from inputData import inputData from connect import connect #test on data preparation inp = inputData() inp.welcome() inp.requestData() inp.represent() #test on connection con = connect(inp.url) content = con.conn() title = con.getHeader(content)