Python config.config.num_classes() Examples
The following are 30
code examples of config.config.num_classes().
You can vote up the ones you like or vote down the ones you don't like,
and go to the original project or source file by following the links above each example.
You may also want to check out all available functions/classes of the module
config.config
, or try the search function
.
.
Example #1
| Source File: data.py From insightface with MIT License | 6 votes |
|
def next_sample(self):
"""Helper function for reading in next sample."""
if self.cur >= len(self.seq):
raise StopIteration
idx = self.seq[self.cur]
self.cur += 1
uv_path = self.uv_file_list[idx]
image_path = self.image_file_list[idx]
uvmap = np.load(uv_path)
img = cv2.imread(image_path)[:,:,::-1]#to rgb
hlabel = uvmap
#print(hlabel.shape)
#hlabel = np.array(header.label).reshape( (self.output_label_size, self.output_label_size, self.num_classes) )
hlabel /= self.input_img_size
return img, hlabel
Example #2
| Source File: eval.py From TorchSeg with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #3
| Source File: eval.py From TorchSeg with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #4
| Source File: eval.py From TorchSeg with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #5
| Source File: eval.py From TorchSeg with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #6
| Source File: eval.py From TorchSeg with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #7
| Source File: eval.py From TorchSeg with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #8
| Source File: eval.py From TorchSeg with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #9
| Source File: eval.py From TorchSeg with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #10
| Source File: eval.py From TorchSeg with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #11
| Source File: eval.py From TorchSeg with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #12
| Source File: eval.py From TorchSeg with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #13
| Source File: eval.py From FNA with Apache License 2.0 | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #14
| Source File: data.py From 1.FaceRecognition with MIT License | 6 votes |
|
def next_sample(self):
"""Helper function for reading in next sample."""
if self.cur >= len(self.seq):
raise StopIteration
idx = self.seq[self.cur]
self.cur += 1
s = self.imgrec.read_idx(idx)
header, img = recordio.unpack(s)
img = mx.image.imdecode(img).asnumpy()
hlabel = np.array(header.label).reshape( (self.num_classes,2) )
if not config.label_xfirst:
hlabel = hlabel[:,::-1] #convert to X/W first
annot = {'scale': config.base_scale}
#ul = np.array( (50000,50000), dtype=np.int32)
#br = np.array( (0,0), dtype=np.int32)
#for i in range(hlabel.shape[0]):
# h = int(hlabel[i][0])
# w = int(hlabel[i][1])
# key = np.array((h,w))
# ul = np.minimum(key, ul)
# br = np.maximum(key, br)
return img, hlabel, annot
Example #15
| Source File: data.py From 1.FaceRecognition with MIT License | 6 votes |
|
def next_sample(self):
"""Helper function for reading in next sample."""
if self.cur >= len(self.seq):
raise StopIteration
idx = self.seq[self.cur]
self.cur += 1
uv_path = self.uv_file_list[idx]
image_path = self.image_file_list[idx]
uvmap = np.load(uv_path)
img = cv2.imread(image_path)[:,:,::-1]#to rgb
hlabel = uvmap
#print(hlabel.shape)
#hlabel = np.array(header.label).reshape( (self.output_label_size, self.output_label_size, self.num_classes) )
hlabel /= self.input_img_size
return img, hlabel
Example #16
| Source File: data_process.py From ecg_pytorch with Apache License 2.0 | 6 votes |
|
def split_data(file2idx, val_ratio=0.1):
'''
划分数据集,val需保证每类至少有1个样本
:param file2idx:
:param val_ratio:验证集占总数据的比例
:return:训练集，验证集路径
'''
data = set(os.listdir(config.train_dir))
val = set()
idx2file = [[] for _ in range(config.num_classes)]
for file, list_idx in file2idx.items():
for idx in list_idx:
idx2file[idx].append(file)
for item in idx2file:
# print(len(item), item)
num = int(len(item) * val_ratio)
val = val.union(item[:num])
train = data.difference(val)
return list(train), list(val)
Example #17
| Source File: data.py From insightface with MIT License | 6 votes |
|
def next_sample(self):
"""Helper function for reading in next sample."""
if self.cur >= len(self.seq):
raise StopIteration
idx = self.seq[self.cur]
self.cur += 1
s = self.imgrec.read_idx(idx)
header, img = recordio.unpack(s)
img = mx.image.imdecode(img).asnumpy()
hlabel = np.array(header.label).reshape( (self.num_classes,2) )
if not config.label_xfirst:
hlabel = hlabel[:,::-1] #convert to X/W first
annot = {'scale': config.base_scale}
#ul = np.array( (50000,50000), dtype=np.int32)
#br = np.array( (0,0), dtype=np.int32)
#for i in range(hlabel.shape[0]):
# h = int(hlabel[i][0])
# w = int(hlabel[i][1])
# key = np.array((h,w))
# ul = np.minimum(key, ul)
# br = np.maximum(key, br)
return img, hlabel, annot
Example #18
| Source File: eval.py From TreeFilter-Torch with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #19
| Source File: eval.py From TreeFilter-Torch with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #20
| Source File: eval.py From TreeFilter-Torch with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #21
| Source File: eval.py From TreeFilter-Torch with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #22
| Source File: eval.py From TreeFilter-Torch with MIT License | 6 votes |
|
def compute_metric(self, results):
hist = np.zeros((config.num_classes, config.num_classes))
correct = 0
labeled = 0
count = 0
for d in results:
hist += d['hist']
correct += d['correct']
labeled += d['labeled']
count += 1
iu, mean_IU, _, mean_pixel_acc = compute_score(hist, correct,
labeled)
result_line = print_iou(iu, mean_pixel_acc,
dataset.get_class_names(), True)
return result_line
Example #23
| Source File: eval.py From TreeFilter-Torch with MIT License | 5 votes |
|
def func_per_iteration(self, data, device):
img = data['data']
label = data['label']
name = data['fn']
pred = self.sliding_eval(img, config.eval_crop_size,
config.eval_stride_rate, device)
hist_tmp, labeled_tmp, correct_tmp = hist_info(config.num_classes,
pred,
label)
results_dict = {'hist': hist_tmp, 'labeled': labeled_tmp,
'correct': correct_tmp}
if self.save_path is not None:
fn = name + '.png'
cv2.imwrite(os.path.join(self.save_path, fn), pred)
logger.info('Save the image ' + fn)
if self.show_image:
colors = self.dataset.get_class_colors()
image = img
clean = np.zeros(label.shape)
comp_img = show_img(colors, config.background, image, clean,
label,
pred)
cv2.imshow('comp_image', comp_img)
cv2.waitKey(0)
return results_dict
Example #24
| Source File: data_process.py From ecg_pytorch with Apache License 2.0 | 5 votes |
|
def count_labels(data, file2idx):
'''
统计每个类别的样本数
:param data:
:param file2idx:
:return:
'''
cc = [0] * config.num_classes
for fp in data:
for i in file2idx[fp]:
cc[i] += 1
return np.array(cc)
Example #25
| Source File: network.py From TorchSeg with MIT License | 5 votes |
|
def get():
return BiSeNet(config.num_classes, None, None)
Example #26
| Source File: eval.py From TorchSeg with MIT License | 5 votes |
|
def func_per_iteration(self, data, device):
img = data['data']
label = data['label']
name = data['fn']
img = cv2.resize(img, (config.image_width, config.image_height),
interpolation=cv2.INTER_LINEAR)
label = cv2.resize(label,
(config.image_width // config.gt_down_sampling,
config.image_height // config.gt_down_sampling),
interpolation=cv2.INTER_NEAREST)
pred = self.whole_eval(img,
(config.image_height // config.gt_down_sampling,
config.image_width // config.gt_down_sampling),
device=device)
hist_tmp, labeled_tmp, correct_tmp = hist_info(config.num_classes,
pred,
label)
results_dict = {'hist': hist_tmp, 'labeled': labeled_tmp,
'correct': correct_tmp}
if self.save_path is not None:
fn = name + '.png'
cv2.imwrite(os.path.join(self.save_path, fn), pred)
logger.info('Save the image ' + fn)
if self.show_image:
colors = self.dataset.get_class_colors
image = img
clean = np.zeros(label.shape)
comp_img = show_img(colors, config.background, image, clean,
label,
pred)
cv2.imshow('comp_image', comp_img)
cv2.waitKey(0)
return results_dict
Example #27
| Source File: network.py From TorchSeg with MIT License | 5 votes |
|
def get():
return BiSeNet(config.num_classes, None, None)
Example #28
| Source File: eval.py From TreeFilter-Torch with MIT License | 5 votes |
|
def func_per_iteration(self, data, device):
img = data['data']
label = data['label']
name = data['fn']
pred = self.sliding_eval(img, config.eval_crop_size,
config.eval_stride_rate, device)
hist_tmp, labeled_tmp, correct_tmp = hist_info(config.num_classes,
pred,
label)
results_dict = {'hist': hist_tmp, 'labeled': labeled_tmp,
'correct': correct_tmp}
if self.save_path is not None:
fn = name + '.png'
cv2.imwrite(os.path.join(self.save_path, fn), pred)
logger.info('Save the image ' + fn)
if self.show_image:
colors = self.dataset.get_class_colors()
image = img
clean = np.zeros(label.shape)
comp_img = show_img(colors, config.background, image, clean,
label,
pred)
cv2.imshow('comp_image', comp_img)
cv2.waitKey(0)
return results_dict
Example #29
| Source File: network.py From TorchSeg with MIT License | 5 votes |
|
def get():
return BiSeNet(config.num_classes, None, None)
Example #30
| Source File: eval.py From TorchSeg with MIT License | 5 votes |
|
def func_per_iteration(self, data, device):
img = data['data']
label = data['label']
name = data['fn']
img = cv2.resize(img, (config.image_width, config.image_height),
interpolation=cv2.INTER_LINEAR)
label = cv2.resize(label,
(config.image_width // config.gt_down_sampling,
config.image_height // config.gt_down_sampling),
interpolation=cv2.INTER_NEAREST)
pred = self.whole_eval(img,
(config.image_height // config.gt_down_sampling,
config.image_width // config.gt_down_sampling),
device=device)
hist_tmp, labeled_tmp, correct_tmp = hist_info(config.num_classes,
pred,
label)
results_dict = {'hist': hist_tmp, 'labeled': labeled_tmp,
'correct': correct_tmp}
if self.save_path is not None:
fn = name + '.png'
cv2.imwrite(os.path.join(self.save_path, fn), pred)
logger.info('Save the image ' + fn)
if self.show_image:
colors = self.dataset.get_class_colors
image = img
clean = np.zeros(label.shape)
comp_img = show_img(colors, config.background, image, clean,
label,
pred)
cv2.imshow('comp_image', comp_img)
cv2.waitKey(0)
return results_dict
