from museotoolbox import datasets, processing
##############################################################################
# Load HistoricalMap dataset
# -------------------------------------------
_, centroid = datasets.load_historical_data(low_res=True, centroid=True)
raster, vector = datasets.load_historical_data(low_res=True)
field = 'Class'
##############################################################################
# Extract label ('Class' field) and groups ('uniquefid' field)
# Compute distanceMatrix with centroid (one point per group)
X, y, groups = processing.extract_ROI(raster, vector, field, 'uniquefid')
distance_matrix, distance_label = processing.get_distance_matrix(
raster, centroid, 'uniquefid')
##############################################################################
# Create CV
# -------------------------------------------
# n_splits will be the number of the least populated class
SLOSGO = SpatialLeaveOneSubGroupOut(distance_thresold=100,
distance_matrix=distance_matrix,
distance_label=distance_label,
random_state=12)
###############################################################################
# .. note::
# Split is made to generate each fold
SLOSGO.get_n_splits(X, y, groups)
Created on Sun Nov 24 12:03:31 2019
@author: nicolas
"""
# -*- coding: utf-8 -*-
import unittest
import os
import numpy as np
from museotoolbox.datasets import load_historical_data
from museotoolbox import cross_validation
from museotoolbox import processing
raster, vector = load_historical_data()
X, y, g = load_historical_data(return_X_y_g=True)
distance_matrix = processing.get_distance_matrix(raster, vector)
n_class = len(np.unique(y, return_counts=True)[1])
smallest_class = np.min(np.unique(y, return_counts=True)[1])
class TestCV(unittest.TestCase):
def test_loo(self):
for split in [False, 1, 2, 5]:
cv = cross_validation.LeaveOneOut(n_repeats=split,
random_state=split,
verbose=split)
if split == False:
assert (cv.get_n_splits(X, y) == np.min(
np.unique(y, return_counts=True)[-1]))
else:
def test_get_distance_matrix(self):
distance_matrix,label = processing.get_distance_matrix(raster,vector,'Class')
assert(label.size== distance_matrix.shape[0])