def load_data(K):
'''
data_path="C://Users//15151//Desktop//xclara.csv"
df=pd.read_csv(data_path)
fig = plt.figure()
sns.FacetGrid(data=df).map(plt.scatter,"V1","V2")
plt.show()
return df.values
'''
t = Transformer()
t.segmenter()
raw_data, name_list = t.numerizer(0.025)
print(hopkins(copy.deepcopy(raw_data)))
d = DR(raw_data, K)
preprocessed_data = d.analyze()
print(hopkins(copy.deepcopy(preprocessed_data)))
return preprocessed_data, name_list
def set_primary_machine_annotation(sample_id, positions):
if positions is None:
sample_update = {'processed': False,
'machine_position_count': None,
'machine_hopkins': None,
'error': False,
'error_string': None}
else:
sample_update = {'processed': True,
'machine_position_count': len(positions),
'machine_hopkins': hopkins(np.array(positions)),
'error': False,
'error_string': None}
samples.update({'_id': sample_id}, {"$set": sample_update}, upsert=False)
if __name__ == '__main__':
ap = argparse.ArgumentParser()
ap.add_argument("-d", "--data", required = False,
help= "path to input data file")
ap.add_argument("-t", "--test", required = False,
help = "test started !")
ap.add_argument("-dr", "--draw", required = False,
help = "draw some graphs!")
ap.add_argument("-hp", "--hopkins", required = False,
help = "find if a set contain cluster")
args = vars(ap.parse_args())
if args["hopkins"]:
X = data.Biofile(args["hopkins"]).table
result = hopkins(X)
print(result)
if args["data"]:
main(args["data"])
if args["draw"]:
test_util.test_plot(args["draw"])
#plot_heatmap_cvx(args["draw"])
#plot_clustering_cvx(args["draw"])
if args["test"]:
if args["test"] == "sample":
test_util.test_result()
elif args["test"] == "Kmeans":
test_util.test_kmeans_result()