#!/usr/bin/env python
import rospy
from jsk_recognition_msgs.msg import DepthErrorResult, PlotData
import numpy as np
import csv
grid = 0.1 # 10 cm
bins = {}
data = PlotData()
def callback(msg):
bin_index = int(msg.true_depth / grid)
if not bins.has_key(bin_index):
bins[bin_index] = []
err = msg.true_depth - msg.observed_depth
bins[bin_index].append((msg, err))
for bin_index, msg_and_errs in bins.items():
if len(msg_and_errs) < 5:
continue
print "Bin: {0}m ~ {1}m".format(grid * bin_index,
grid * (bin_index + 1))
print " Sample:", len(msg_and_errs)
errs = [err for (msg, err) in msg_and_errs]
mean = np.mean(errs, axis=0)
stddev = np.std(errs, axis=0)
print " Mean:", mean
print " Stddev:", stddev
data.xs.append(msg.true_depth)
#!/usr/bin/env python
from jsk_recognition_msgs.msg import PlotData
import numpy as np
import rospy
import math
if __name__ == "__main__":
rospy.init_node("sample_2d_plot")
pub = rospy.Publisher("~output", PlotData)
r = rospy.Rate(10)
offset = 0
while not rospy.is_shutdown():
msg = PlotData()
msg.xs = np.arange(0, 5, 0.1)
msg.ys = [math.sin(x + offset) for x in msg.xs]
pub.publish(msg)
offset = offset + 0.1
r.sleep()
#!/usr/bin/env python
from jsk_recognition_msgs.msg import PlotData
import numpy as np
import rospy
from numpy.random import *
import math
if __name__ == "__main__":
rospy.init_node("sample_2d_plot")
pub = rospy.Publisher("~output", PlotData)
r = rospy.Rate(10)
offset = 0
while not rospy.is_shutdown():
msg = PlotData()
msg.xs = randn(50)
msg.ys = randn(50)
pub.publish(msg)
offset = offset + 0.1
r.sleep()
#!/usr/bin/env python
from jsk_recognition_msgs.msg import PlotData
import numpy as np
import rospy
import math
from random import random
if __name__ == "__main__":
rospy.init_node("sample_2d_plot")
pub = rospy.Publisher("~output", PlotData)
r = rospy.Rate(10)
offset = 0
while not rospy.is_shutdown():
msg = PlotData()
msg.xs = np.arange(0, 5, 0.1)
msg.ys = [math.sin(x + offset) for x in msg.xs]
msg.label = "sample data"
if random() < 0.5:
msg.type = PlotData.LINE
else:
msg.type = PlotData.SCATTER
msg.fit_line = random() < 0.5
msg.fit_line_ransac = random() < 0.5
pub.publish(msg)
offset = offset + 0.1
r.sleep()
