Sphere
[1]:
%matplotlib inline
[2]:
# Not really required
import sys
sys.path.insert(0, '../../..')
[3]:
from pyLDLE2 import datasets
from pyLDLE2 import buml_, util_
matplotlib.get_backend() = module://matplotlib_inline.backend_inline
[4]:
save_dir_root = '../data/pyLDLE2/sphere/'
[5]:
X, labelsMat, ddX = datasets.Datasets().sphere()
X.shape = (10000, 3)
[6]:
# The supplied options would override the default options
buml_obj = buml_.BUML(local_opts={'algo':'LPCA'},
intermed_opts={'eta_min': 5},
vis_opts={'c': labelsMat[:,0], 'save_dir': save_dir_root},
verbose=True, debug=True)
local_opts['k_nn0'] = 700 is created.
Options provided:
local_opts:
{
"Atilde_method": "LDLE_1",
"N": 100,
"U_method": "k_nn",
"algo": "LPCA",
"alpha": 1,
"debug": true,
"delta": 0.9,
"gl_type": "unnorm",
"k": 28,
"k_nn": 49,
"k_nn0": 700,
"k_tune": 7,
"lambda1_decay": 0.75,
"lambda1_init": 8,
"lambda1_min": 0.001,
"max_iter": 300,
"max_sparsity": 0.9,
"metric": "euclidean",
"n_proc": 32,
"p": 0.99,
"power": 5,
"pp_n_thresh": 32,
"radius": 0.5,
"reg": 0.0,
"scale_by": "gamma",
"tau": 50,
"to_postprocess": true,
"verbose": true
}
intermed_opts:
{
"algo": "best",
"debug": true,
"eta_max": 25,
"eta_min": 5,
"len_S_thresh": 256,
"local_algo": "LPCA",
"metric": "euclidean",
"n_proc": 32,
"n_times": 4,
"verbose": true
}
global_opts:
{
"add_dim": false,
"align_transform": "rigid",
"align_w_parent_only": true,
"alpha": 0.3,
"beta": {
"align": null,
"repel": 1
},
"color_tear": true,
"compute_error": false,
"debug": true,
"eps": 1e-08,
"err_tol": 0.0001,
"far_off_points_type": "reuse_fixed",
"init_algo_name": "procrustes",
"k": 28,
"max_internal_iter": 100,
"max_iter": 20,
"metric": "euclidean",
"n_proc": 32,
"n_repel": 0,
"nu": 3,
"patience": 5,
"refine_algo_name": "rgd",
"repel_by": 0.0,
"to_tear": true,
"verbose": true,
"vis_before_init": false
}
[7]:
buml_obj.fit(X=X)
Computing local views using LPCA
local_param: all 10000 points processed...
max distortion is 1.004248
Done.
##############################
Time elapsed from last time log: 1.2 seconds
Total time elapsed: 1.2 seconds
##############################
Posprocessing local parameterizations.
#Param replaced: 0, max distortion: 1.004248
Max local distortion after postprocessing: 1.0042477110911947
Done.
##############################
Time elapsed from last time log: 2.5 seconds
Total time elapsed: 3.7 seconds
##############################
Max local distortion = 1.0042477110911947
Constructing intermediate views.
eta = 2.
#non-empty views with sz < 2 = 10000
#nodes in views with sz < 2 = 10000
Costs computed when eta = 2.
##############################
Time elapsed from last time log: 10.9 seconds
Total time elapsed: 12.0 seconds
##############################
ctr=5624, total_len_S=16810, avg_len_S=2.989
Remaining #nodes in views with sz < 2 = 0
Done with eta = 2.
##############################
Time elapsed from last time log: 60.6 seconds
Total time elapsed: 72.6 seconds
##############################
eta = 3.
#non-empty views with sz < 3 = 3324
#nodes in views with sz < 3 = 6648
Costs computed when eta = 3.
##############################
Time elapsed from last time log: 4.2 seconds
Total time elapsed: 76.8 seconds
##############################
ctr=3679, total_len_S=63804, avg_len_S=17.343
Remaining #nodes in views with sz < 3 = 0
Done with eta = 3.
##############################
Time elapsed from last time log: 199.2 seconds
Total time elapsed: 276.0 seconds
##############################
eta = 4.
#non-empty views with sz < 4 = 1286
#nodes in views with sz < 4 = 3858
Costs computed when eta = 4.
##############################
Time elapsed from last time log: 2.4 seconds
Total time elapsed: 278.4 seconds
##############################
ctr=2594, total_len_S=65760, avg_len_S=25.351
Remaining #nodes in views with sz < 4 = 0
Done with eta = 4.
##############################
Time elapsed from last time log: 85.2 seconds
Total time elapsed: 363.6 seconds
##############################
eta = 5.
#non-empty views with sz < 5 = 612
#nodes in views with sz < 5 = 2448
Costs computed when eta = 5.
##############################
Time elapsed from last time log: 1.8 seconds
Total time elapsed: 365.4 seconds
##############################
ctr=1918, total_len_S=55970, avg_len_S=29.181
Remaining #nodes in views with sz < 5 = 0
Done with eta = 5.
##############################
Time elapsed from last time log: 44.3 seconds
Total time elapsed: 409.7 seconds
##############################
Pruning and cleaning up.
Done.
##############################
Time elapsed from last time log: 1.2 seconds
Total time elapsed: 410.9 seconds
##############################
After clustering, max distortion is 1.012415
Using rigid transforms for alignment.
Computing laterations scores for overlaps b/w intermed views
Done
##############################
Time elapsed from last time log: 1.3 seconds
Total time elapsed: 1.4 seconds
##############################
Computing a lateration.
Seq of intermediate views and their predecessors computed.
No. of connected components = 1
Done.
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 1.4 seconds
##############################
Computing initial embedding using: procrustes algorithm
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 1.4 seconds
##############################
Initial alignment of 361 views completed
Initial alignment of 722 views completed
Initial alignment of 1083 views completed
Initial alignment of 1444 views completed
Embedding initialized.
##############################
Time elapsed from last time log: 1.6 seconds
Total time elapsed: 2.9 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 0
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 17.4 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 10.9 seconds
Total time elapsed: 28.3 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 1
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 43.7 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 20.9 seconds
Total time elapsed: 64.6 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 2
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 74.4 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 11.4 seconds
Total time elapsed: 85.8 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 3
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 95.4 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 10.6 seconds
Total time elapsed: 106.0 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 4
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 115.9 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 11.3 seconds
Total time elapsed: 127.2 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 5
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 137.3 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 17.1 seconds
Total time elapsed: 154.3 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 6
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 163.1 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 18.3 seconds
Total time elapsed: 181.3 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 7
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 192.3 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 17.5 seconds
Total time elapsed: 209.7 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 8
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 219.6 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 14.1 seconds
Total time elapsed: 233.7 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 9
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 243.9 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 10.5 seconds
Total time elapsed: 254.4 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 10
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 263.7 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 11.4 seconds
Total time elapsed: 275.2 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 11
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 283.5 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 12.4 seconds
Total time elapsed: 296.0 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 12
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 302.8 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 25.8 seconds
Total time elapsed: 328.7 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 13
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 338.8 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 17.3 seconds
Total time elapsed: 356.0 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 14
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 364.8 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 13.1 seconds
Total time elapsed: 378.0 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 15
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 387.6 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 10.9 seconds
Total time elapsed: 398.5 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 16
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 408.1 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 16.4 seconds
Total time elapsed: 424.5 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 17
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 433.5 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 25.1 seconds
Total time elapsed: 458.6 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 18
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 469.5 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 23.7 seconds
Total time elapsed: 493.2 seconds
##############################
Refining with rgd algorithm for 100 iterations.
Refinement iteration: 19
##############################
Time elapsed from last time log: 0.0 seconds
Total time elapsed: 501.6 seconds
##############################
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Descent starts
Done.
##############################
Time elapsed from last time log: 19.8 seconds
Total time elapsed: 521.4 seconds
##############################
Computing error.
min and max weights: 1.0 1.0
Computing Pseudoinverse of a matrix of L of size 10000 + 1446 multiplied with B
Computing svd
Done
Alignment error: 2.496967
##############################
Time elapsed from last time log: 8.7 seconds
Total time elapsed: 530.1 seconds
##############################
[7]:
array([[-0.13668623, -0.37035817],
[-0.14814647, -0.37076796],
[-0.137852 , -0.36053965],
...,
[ 0.12887644, 0.37633278],
[ 0.13743101, 0.38798857],
[ 0.1265057 , 0.38635904]])
[8]:
%matplotlib inline
buml_obj.GlobalViews.vis_embedding(buml_obj.GlobalViews.y_final, buml_obj.vis, buml_obj.vis_opts,
buml_obj.GlobalViews.color_of_pts_on_tear_final,
title='Final global embedding', )
[9]:
util_.save(save_dir_root, 'lpca.dat', [X, labelsMat, buml_obj])
[ ]: