for i in xrange(1,501):
filename = '%04d' %i + '_input_pred.gif'
f.write(filename+'\n')
# upload on a webpage
t = Table()
srcpath = './gifs/'
image_set_file = 'path_gifs.txt'
with open(image_set_file) as f:
image_index = [x.strip() for x in f.readlines()]
trajnum = len(image_index) // 10
for r in range(trajnum):
idx = r
# if r == 0:
# r = TableRow(isHeader = True)
# else:
r = TableRow()
for e in range(10):
idx2 = idx * 10 + e + 1
e = Element()
tpath = srcpath + '%04d' % idx2 + '_input_pred.gif'
print tpath
e.addImg(tpath)
r.addElement(e)
t.addRow(r)
tw = TableWriter(t, 'out')
tw.write()
import sys
sys.path.append('/home/rohit/Software/vis/PyHTMLWriter/src');
from Element import Element
from TableRow import TableRow
from Table import Table
from TableWriter import TableWriter
import numpy as np
t = Table()
for r in range(100):
if r == 0:
r = TableRow(isHeader = True)
else:
r = TableRow()
for e in range(10):
e = Element()
pose = np.array([[587, 569, 490, 535, 621, 630, 512, 490, 488, 483, 576, 479, 457, 522, 550, 571, 491],[561, 447, 411, 393, 430, 537, 402, 298, 275, 201, 394, 377, 305, 289, 348, 367, 376]]).transpose().tolist()
e.addImg("http://sun.pc.cs.cmu.edu/~rohit/Public/Images/himym_chairs.jpg", poses=[pose], width=500, imsize=[1280,768])
r.addElement(e)
t.addRow(r)
tw = TableWriter(t, 'out')
tw.write()
r.addElement(Element('Sno'))
r.addElement(Element('Method'))
r.addElement(Element('Query/Top Match'))
r.addElement(Element('Matches...'))
for i in range(500):
fpath = os.path.join(respath, str(i + 1) + '.txt');
fpath2 = os.path.join(respath_f, str(i + 2) + '.txt');
imnames, scores = readScores(fpath)
imnames2, scores2 = readScores2(fpath2)
if len(imnames) == 0 or len(imnames2) == 0:
continue
r = TableRow(rno=i)
r.addElement(Element('Background'))
for j in range(20):
e = Element()
e.addImg(os.path.join(imgsdir, imnames[j]),
overlay_path=os.path.join(segdir, imnames[j]))
e.addTxt(scores[j])
r.addElement(e)
t.addRow(r)
r = TableRow(rno=i)
r.addElement(Element('Full Image'))
for j in range(20):
e = Element()
e.addImg(os.path.join(imgsdir, imnames2[j]))
e.addTxt(scores2[j])
r.addElement(e)
t.addRow(r)
tw = TableWriter(t, 'out')
tw.write()
imgsdir = 'http://10.1.94.128:8000/~rgirdhar/memex/dataset/0001_Backpage/Images/'
imgslistpath = '/memexdata/Dataset/processed/0001_Backpage/Images/lists/Images.txt'
def matches_prev(a, b):
if a[:-6] == b[:-6]:
return True
else:
return False
f = open(imgslistpath)
imgslist = f.read().splitlines()
f.close()
t = Table()
r = TableRow()
last_name = ''
for i in range(5000):
e = Element()
if not matches_prev(imgslist[i], last_name):
t.addRow(r)
r = TableRow()
e.addImg(os.path.join(imgsdir, imgslist[i]))
r.addElement(e)
last_name = imgslist[i]
t.addRow(r)
tw = TableWriter(t, 'out')
tw.write()
def export(args, model, dataloader, dataset):
# remove head
model.top_layer = None
model.classifier = nn.Sequential(*list(model.classifier.children())[:-1])
# get the features for the whole dataset
features, idxs, pos1 = dc_main.compute_features(dataloader, model, len(dataset), args)
idxs = idxs[np.argsort(idxs)]
features = features[np.argsort(idxs)]
if args.group > 1:
args.group = args.ep_length - args.traj_length + 1
# clustering algorithm to use
deepcluster = clustering.__dict__[args.clustering](args.nmb_cluster, group=args.group)
# cluster the features
clustering_loss = deepcluster.cluster(features, verbose=args.verbose)
centroids = deepcluster.clus.centroids
# centroids = faiss.vector_float_to_array(deepcluster.clus.get_means_and_variances)
# centroids = centroids.reshape(nmb_cluster, 256)
# import pdb; pdb.set_trace()
# self_index = faiss.IndexFlatL2(centroids.shape[1]) # build the index
# self_index.add(centroids)
# self_dists = self_index.search(centroids, centroids.shape[0])
_, (mean, std), _, _ = vis_utils.make_transform(args.data)
model.features = model.features.module
c_mean, c_cov, c_var = get_means_and_variances(deepcluster, features, args)
resume = args.resume if len(args.resume) > 0 else args.exp
out = {
'state_dict': model.state_dict(), 'centroids': centroids,
'pca_path': resume + '.pca',
'mean': mean, 'std': std,
# 'cluster_mean': c_mean, 'cluster_cov': c_cov,
'clus': deepcluster.clus,
}
if args.export > 0:
faiss.write_VectorTransform(deepcluster.mat, resume + '.pca')
torch.save(out,
resume + '.clus')
out['pca'] = deepcluster.mat
T = args.traj_length
pos = pos1
if sum(sum(pos)) == 0:
meta = torch.load('%s/meta.dict' % args.data)
pos = np.array(meta['pos'])
pos_idx = np.arange(pos.shape[0]*pos.shape[1])
pos_idx = pos_idx.reshape(pos.shape[0], pos.shape[1])[:, T-1:]
pos_idx = pos_idx.reshape(pos_idx.shape[0] * pos_idx.shape[1])
pos = pos.reshape(pos.shape[0]*pos.shape[1], pos.shape[2])
else:
meta = torch.load('/data3/ajabri/vizdoom/single_env_hard_fixed1/0/meta.dict')
# import pdb; pdb.set_trace()
sz = 30
from scipy.ndimage.filters import gaussian_filter
def get_obj_masks(objs):
out = np.zeros((3, sz, sz))
for o in objs[0]:
# import pdb; pdb.set_trace()
x, y = o
x, y = int((x - x0)/x1 *sz), int((y-y0)/y1 * sz)
out[:, x:x+1, y:y+1] = 1
return out
def get_mask_from_coord(coord):
import matplotlib.cm as cm
x, y, a = coord
x, y = int(x), int(y)
out = np.zeros((3, sz, sz))
out[:, x, y] = cm.jet(a)[:3]
return out
# import pdb; pdb.set_trace()
# sorted_self_dists = np.argsort(self_dists[0][:, 1])[::-1]
# sorted_self_dists = np.argsort(self_dists[0].sum(axis=-1))[::-1]
smoother1 = models.mini_models.GaussianSmoothing(3, 5, 5)
smoother2 = models.mini_models.GaussianSmoothing(3, 7, 5)
smoother3 = models.mini_models.GaussianSmoothing(3, 7, 7)
smoother4 = models.mini_models.GaussianSmoothing(3, 9, 7)
exp_name = args.resume.split('/')[-2] if args.resume != '' else args.exp.split('/')[-1]
out_root = '%s/%s' % (args.export_path, exp_name)
# import pdb; pdb.set_trace()
if not os.path.exists(out_root):
os.makedirs(out_root)
table = Table()
num_show = 8
sorted_variance = np.argsort(c_var)[::-1]
sorted_sizes = np.argsort([len(ll) for ll in deepcluster.images_dists])[::-1]
# import pdb; pdb.set_trace()
for c, clus_idx in enumerate(sorted_sizes):
# for c, clus_idx in enumerate(sorted_variance):
# for c, clus_idx in enumerate(sorted_self_dists):
l = deepcluster.images_dists[clus_idx]
if len(l) == 0:
continue
ll = [ii[0] for ii in sorted(l, key=lambda x: x[1])[::-1]][:num_show//2]
ll += [ii[0] for ii in random.sample(l, min(num_show//2, len(l)))]
l = [ii[0] for ii in l]
row = TableRow(rno=c)
e = Element()
e.addTxt('size: %s <br>variance: %s' % (len(deepcluster.images_dists[clus_idx]), c_var[clus_idx]))
row.addElement(e)
# import pdb; pdb.set_trace()
## MAP
poo = []
for t in range(T):
poo += [pos[np.array(l) - t]]
posum = env.make_pose_map(np.concatenate(poo), meta['objs'][0], sz=sz)
# posum *= 255.0
# vis.image((posum*255.).astype(np.uint8), opts=dict(width=300, height=300))
# vis.image(gaussian_filter((posum*255.), sigma=1).astype(np.uint8), opts=dict(width=300, height=300))
# gifname = '%s/%s_%s.png' % (exp_name, c, 'map')
gifname = '%s_%s.png' % (c, 'map')
gifpath = '%s/%s' % (out_root, gifname)
imageio.imwrite(gifpath,
cv2.resize((posum*255.).astype(np.uint8).transpose(1, 2, 0),
(0,0), fx=5, fy=5, interpolation = cv2.INTER_AREA))
e = Element()
e.addImg(gifname, width=180)
row.addElement(e)
## EXEMPLARS
for iii, i in enumerate(ll):
# import pdb; pdb.set_trace()
imgs = vis_utils.unnormalize_batch(dataset[i][0], mean, std)
# vis.images(imgs, opts=dict(title=f"{c} of length {len(l)}"))
# vis.images(smoother1(torch.Tensor(imgs)).numpy(), opts=dict(title=f"{c} of length {len(l)}"))
# vis.images(smoother2(torch.Tensor(imgs)).numpy(), opts=dict(title=f"{c} of length {len(l)}"))
# vis.images(smoother3(torch.Tensor(imgs)).numpy(), opts=dict(title=f"{c} of length {len(l)}"))
# vis.images(smoother4(torch.Tensor(imgs)).numpy(), opts=dict(title=f"{c} of length {len(l)}"))
# gifname = '%s/%s_%s.gif' % (exp_name, c, i)
gifname = '%s_%s.gif' % (c, i)
gifpath = '%s/%s' % (out_root, gifname)
vis_utils.make_gif_from_tensor(imgs.astype(np.uint8), gifpath)
e = Element()
if iii < num_show // 2:
e.addTxt('rank %i<br>' % iii)
else:
e.addTxt('random<br>')
e.addImg(gifname, width=128)
row.addElement(e)
## EXEMPLARS
# import visdom
# vis = visdom.Visdom(port=8095, env='main', use_incoming_socket=False)
# vis.images(vis_utils.unnormalize_batch(
# np.stack([dataset[iii][0][0] for iii in range(-100, -50)]), mean, std
# ))
# import pdb; pdb.set_trace()
# gl = np.array(l).reshape(-1, args.group)
# if args.group > 10:
# exemplars = gl[random.sample(list(range(gl.shape[0])), 4)]
# else:
# exemplars = gl[random.sample(list(range(gl.shape[0])), 10)]
# for iii, i in enumerate(exemplars):
# # import pdb; pdb.set_trace()
# # imgs = vis_utils.unnormalize_batch(dataset[i][0], mean, std)
# imgs = np.stack([dataset[_idx][0][0] for _idx in i])
# imgs = vis_utils.unnormalize_batch(imgs, mean, std)
# # import pdb; pdb.set_trace()
# # imgs = vis_utils.unnormalize_batch(dataset[i][0], mean, std)
# # vis.images(imgs, opts=dict(title=f"{c} of length {len(l)}"))
# # vis.images(smoother1(torch.Tensor(imgs)).numpy(), opts=dict(title=f"{c} of length {len(l)}"))
# # vis.images(smoother2(torch.Tensor(imgs)).numpy(), opts=dict(title=f"{c} of length {len(l)}"))
# # vis.images(smoother3(torch.Tensor(imgs)).numpy(), opts=dict(title=f"{c} of length {len(l)}"))
# # vis.images(smoother4(torch.Tensor(imgs)).numpy(), opts=dict(title=f"{c} of length {len(l)}"))
# # gifname = '%s/%s_%s.gif' % (exp_name, c, i)
# gifname = '%s_%s.gif' % (c, i[0])
# gifpath = '%s/%s' % (out_root, gifname)
# vis_utils.make_gif_from_tensor(imgs.astype(np.uint8), gifpath)
# e = Element()
# if iii < num_show // 2:
# e.addTxt('rank %i<br>' % iii)
# else:
# e.addTxt('random<br>')
# e.addImg(gifname, width=128)
# row.addElement(e)
table.addRow(row)
tw = TableWriter(table, '%s/%s' % (args.export_path, exp_name), rowsPerPage=min(args.nmb_cluster,100))
tw.write()
# import pdb; pdb.set_trace()
return out
t = Table()
srcpath = 'http://ladoga.graphics.cs.cmu.edu/xiaolonw/affordance_TBBT/pose_det_vis/'
image_set_file = 'paths.txt'
with open(image_set_file) as f:
image_index = [x.strip() for x in f.readlines()]
trajnum = len(image_index)
for r in range(trajnum):
idx = r
if r == 0:
r = TableRow(isHeader = True)
else:
r = TableRow()
now_path = image_index[idx]
now_path = srcpath + now_path
e = Element()
e.addTxt(image_index[idx])
r.addElement(e)
for e in range(30):
idx2 = e * 5 + 1
e = Element()
tpath = now_path + '{0:07d}'.format(idx2) + '.jpg'
e.addImg(tpath)
r.addElement(e)
t.addRow(r)
tw = TableWriter(t, 'out')
tw.write()
def writeHTML(video_ls, output_folder_name, genre):
length = len(video_ls)
t = Table()
print('length : ', length)
for r in range(length):
count = r
if r == 0:
r = TableRow(isHeader=True)
a1 = Element()
a1.addTxt("No.")
r.addElement(a1)
a2 = Element()
a2.addTxt("Viedo_ID")
r.addElement(a2)
a3 = Element()
a3.addTxt("Genre")
r.addElement(a3)
a4 = Element()
a4.addTxt("Handscore")
r.addElement(a4)
a4 = Element()
a4.addTxt("Is_beauty")
r.addElement(a4)
a4 = Element()
a4.addTxt("Is_cartoon")
r.addElement(a4)
a5 = Element()
a5.addTxt("Thumbnail_1")
r.addElement(a5)
a6 = Element()
a6.addTxt("Thumbnail_2")
r.addElement(a6)
a7 = Element()
a7.addTxt("Thumbnail_3")
r.addElement(a7)
a8 = Element()
a8.addTxt("Thumbnail_4")
r.addElement(a8)
a9 = Element()
a9.addTxt("Link")
r.addElement(a9)
t.addRow(r)
continue
else:
r = TableRow()
# print(count)
# print(v_id[count-1])
# print(label_json[v_id[count-1]])
# print(genre[str(label_json[v_id[count-1]]) ])
a = Element() # no.
a.addTxt("No." + str(count))
r.addElement(a)
b = Element() # video id
b.addTxt(video_ls[count - 1][0])
r.addElement(b)
c = Element() # genre
c.addTxt(genre)
r.addElement(c)
c = Element() # score
c.addTxt(video_ls[count - 1][1])
r.addElement(c)
c = Element() # score
c.addTxt(video_ls[count - 1][2])
r.addElement(c)
c = Element() # score
c.addTxt(video_ls[count - 1][0])
r.addElement(c)
for i in range(4): # 4 thumbnails
f = Element()
f.addImg("https://img.youtube.com/vi/" + video_ls[count - 1][0] +
"/" + str(i) + ".jpg")
r.addElement(f)
g = Element("click me") # link
g.addLink("https://www.youtube.com/watch?v=" + video_ls[count - 1][0])
r.addElement(g)
t.addRow(r)
tw = TableWriter(t, output_folder_name)
tw.write()
from Table import Table
from TableWriter import TableWriter
import numpy as np
t = Table()
for r in range(100):
if r == 0:
r = TableRow(isHeader=True)
else:
r = TableRow()
for e in range(10):
e = Element()
pose = np.array([[
587, 569, 490, 535, 621, 630, 512, 490, 488, 483, 576, 479, 457,
522, 550, 571, 491
],
[
561, 447, 411, 393, 430, 537, 402, 298, 275, 201,
394, 377, 305, 289, 348, 367, 376
]]).transpose().tolist()
e.addImg(
"http://sun.pc.cs.cmu.edu/~rohit/Public/Images/himym_chairs.jpg",
poses=[pose],
width=500,
imsize=[1280, 768])
r.addElement(e)
t.addRow(r)
tw = TableWriter(t, 'out')
tw.write()
import sys
sys.path.append('/home/dragon123/Affordances/PyHTMLWriter/src');
from Element import Element
from TableRow import TableRow
from Table import Table
from TableWriter import TableWriter
t = Table()
for i in range(1):
if i == 0:
r = TableRow(isHeader = True)
else:
r = TableRow(rno = i)
for e in range(10):
e = Element()
e.setDrawCheck()
e.addImg('eiffeltower.jpg', bboxes=[[900,500,50,34],[100,100,100,100]])
r.addElement(e)
t.addRow(r)
r = TableRow(100)
for e in range(10):
e = Element()
e.setDrawCheck()
e.addImg('eiffeltower.jpg', overlay_path='../taj.jpg')
r.addElement(e)
t.addRow(r)
tw = TableWriter(t, 'out')
tw.write()
import sys
sys.path.append('/home/rohit/Software/utils/PyHTMLWriter/src');
from Element import Element
from TableRow import TableRow
from Table import Table
from TableWriter import TableWriter
t = Table()
print "NOTE that this is buggy still... will make images at the std size of canvas, and will put the square at absolute values mentioned"
for r in range(100):
if r == 0:
r = TableRow(isHeader = True)
else:
r = TableRow()
for e in range(10):
e = Element()
# X, Y, W, H
e.addImg("http://upload.wikimedia.org/wikipedia/commons/d/db/A_small_forest_clearing_-_geograph.org.uk_-_540354.jpg", [], [20,20,100,150])
r.addElement(e)
t.addRow(r)
tw = TableWriter(t, 'out')
tw.write()
r.addElement(Element('Query/Top Match'))
r.addElement(Element('Matches...'))
for i in range(500):
fpath = os.path.join(respath,
str(i + 1) + '.txt')
fpath2 = os.path.join(respath_f,
str(i + 2) + '.txt')
imnames, scores = readScores(fpath)
imnames2, scores2 = readScores2(fpath2)
if len(imnames) == 0 or len(imnames2) == 0:
continue
r = TableRow(rno=i)
r.addElement(Element('Background'))
for j in range(20):
e = Element()
e.addImg(os.path.join(imgsdir, imnames[j]),
overlay_path=os.path.join(segdir, imnames[j]))
e.addTxt(scores[j])
r.addElement(e)
t.addRow(r)
r = TableRow(rno=i)
r.addElement(Element('Full Image'))
for j in range(20):
e = Element()
e.addImg(os.path.join(imgsdir, imnames2[j]))
e.addTxt(scores2[j])
r.addElement(e)
t.addRow(r)
tw = TableWriter(t, 'out')
tw.write()
