if truncate:
Y_samples = Y_samples[good_inds, :]
# outfile = 'bh_da_init_' + infile + '_out.txt'
# betafile = 'bh_da_init_' + infile + '_betas.txt'
print(pc_data.shape)
# DROPOUT
## sums = np.sum(pc_data**2, axis=1)
## print(min(sums))
## pc_data = pc_data[sums > .1, :]
## print(pc_data.shape)
embedded, betas, orig_densities, emb_densities = run_bh_tsne(
pc_data,
initial_dims=pc_data.shape[1],
theta=0.3,
verbose=True,
perplexity=50,
max_iter=max_iter,
use_pca=False,
Y_samples=Y_samples,
weight=1.)
np.savetxt(file_root.format(outdir, infile, 'out'), embedded)
np.savetxt(file_root.format(outdir, infile, 'betas'), betas)
np.savetxt(file_root.format(outdir, infile, 'marg_origD'), orig_densities)
np.savetxt(file_root.format(outdir, infile, 'marg_embD'), emb_densities)
## sums = np.sum(pc_data**2, axis=1)
## print(min(sums))
## pc_data = pc_data[sums > .1, :]
## print(pc_data.shape)
Y_file = 'bh_' + infile + '_out.txt'
Y_samples = np.loadtxt(Y_file)
# DEBUG PRESENCE OF INITIAL SAMPLES
# weights = [1.0, 2.0, 3.0]
for i, w in enumerate(weights):
embedded, betas, orig_densities, emb_densities = bh_da_sne_init.run_bh_tsne(
pc_data,
initial_dims=pc_data.shape[1],
theta=0.3,
thresh=1000.0,
verbose=True,
perplexity=30,
max_iter=max_iter,
use_pca=False,
Y_samples=Y_samples,
weight=w)
print embedded.shape, betas.shape
np.savetxt(file_root.format(infile, 'out', i), embedded)
np.savetxt(file_root.format(infile, 'betas', i), betas)
np.savetxt(file_root.format(infile, 'marg_origD', i), orig_densities)
np.savetxt(file_root.format(infile, 'marg_embD', i), emb_densities)
if (initY is not None):
Y_samples = np.loadtxt(initY)
max_iter = 500
print transformed.shape
if (initY is not None):
print Y_samples.shape
N, D = transformed.shape
if sub:
sub_sz = int(subsample * N)
indices = np.random.choice(N, sub_sz, replace=False)
transformed = transformed[indices, :]
np.savetxt(pcafile + '.txt', transformed)
embedded, betas = bh_da_sne_init.run_bh_tsne(transformed,
initial_dims=transformed.shape[1],
theta=0.3,
thresh=1.0,
verbose=True,
perplexity=30,
max_iter=max_iter,
use_pca=False,
Y_samples=Y_samples)
print embedded.shape, betas.shape
np.savetxt(outfile, embedded)
np.savetxt(betafile, betas)