Device chosen is "GeForce GTX 1070"
Device has 15 multi processors and compute capability 6.1
Max threads per block supported are 1024

Reading dataset and labels...
Done.

Device memory allocation wall clock time = 0.094704

calculate_kernel_matrix_kernel called with:
dimBlock.x = 32, dimBlock.y = 32
dimGrid.x = 32, dimGrid.y = 32

calculate_denominator called with:
dimBlock.x = 1024, dimBlock.y = 1
dimGrid.x = 1, dimGrid.y = 1

shift_points_kernel called with:
dimBlock.x = 32, dimBlock.y = 32
dimGrid.x = 32, dimGrid.y = 1

Recursion n. 0, error	212.611066 
Recursion n. 1, error	51.768217 
Recursion n. 2, error	18.321997 
Recursion n. 3, error	7.902559 
Recursion n. 4, error	3.830385 
Recursion n. 5, error	1.990884 
Recursion n. 6, error	1.077207 
Recursion n. 7, error	0.596253 
Recursion n. 8, error	0.334476 
Recursion n. 9, error	0.189225 
Recursion n. 10, error	0.107681 
Recursion n. 11, error	0.061545 
Recursion n. 12, error	0.035299 
Recursion n. 13, error	0.020304 
Recursion n. 14, error	0.011708 
Recursion n. 15, error	0.006766 
Recursion n. 16, error	0.003918 
Recursion n. 17, error	0.002273 
Recursion n. 18, error	0.001321 
Recursion n. 19, error	0.000769 
Recursion n. 20, error	0.000448 
Recursion n. 21, error	0.000262 
Recursion n. 22, error	0.000153 
Recursion n. 23, error	0.000090 

Copying between device and host wall clock time = 0.046973

Total number of recursions = 23
Mean Shift wall clock time = 0.713939