MCPcopy Create free account

hub / github.com/buaacxf/VIPTR / functions

Functions373 in github.com/buaacxf/VIPTR

↓ 17 callersMethod__init__
(self, embed_dim, num_heads, value_factor=1)
modules/VIPTRv1-T_ch.py:213
↓ 17 callersMethod__init__
(self, embed_dim, num_heads, value_factor=1)
modules/VIPTRv1.py:213
↓ 17 callersMethod__init__
(self, embed_dim, num_heads, value_factor=1)
modules/VIPTRv2.py:208
↓ 17 callersFunctionrad
rad
dataload/dataAug.py:232
↓ 16 callersMethod__init__
(self, embed_dim, num_heads, value_factor=1)
modules/VIPTRv2T_ch.py:208
↓ 9 callersMethod__init__
(self, in_features, hidden_features=None, out_features=None
modules/SVTR.py:89
↓ 9 callersMethodencode
convert text-label into text-index. input: text: text labels of each image. [batch_size] batch_max_length: max length
utils.py:19
↓ 6 callersMethod__init__
(self, input_channel, output_channel, num_iteration, kernel_size, pad)
modules/feature_extraction.py:68
↓ 6 callersFunctionconv3x3_block
3x3 convolution with padding
modules/stn_head.py:11
↓ 6 callersMethoddecode
convert text-index into text-label.
utils.py:40
↓ 5 callersMethod__init__
(self, step_each_epoch, step_size, warmup_epoch=0,
optimizer.py:25
↓ 4 callersMethod_make_layer
(self, block, planes, blocks, stride=1)
modules/feature_extraction.py:194
↓ 4 callersFunctionflag
flag
dataload/dataAug.py:63
↓ 4 callersFunctionhierarchical_dataset
select_data='/' contains all sub-directory of root directory
dataset.py:103
↓ 4 callersMethodreset_parameters
(self)
modules/VIPTRv1-T_ch.py:228
↓ 4 callersMethodreset_parameters
(self)
modules/VIPTRv2T_ch.py:223
↓ 4 callersMethodreset_parameters
(self)
modules/VIPTRv1.py:228
↓ 4 callersMethodreset_parameters
(self)
modules/VIPTRv2.py:223
↓ 3 callersMethod__bilinear_interp
(x, y, v11, v12, v21, v22)
dataload/aug/warp_mls.py:20
↓ 3 callersMethod__init__
(self, input_size, hidden_size, output_size)
modules/sequence_modeling.py:66
↓ 3 callersMethodgenerate
(self)
dataload/aug/warp_mls.py:23
↓ 3 callersFunctionvalidation
validation or evaluation
test_benchmark.py:80
↓ 2 callersFunctionVIPTRv1
(opt)
modules/VIPTRv1.py:977
↓ 2 callersFunctionVIPTRv2T_CH
(opt)
modules/VIPTRv2T_ch.py:861
↓ 2 callersMethod__init__
Generate P_hat and inv_delta_C for later
modules/transformation.py:89
↓ 2 callersMethod_char_to_onehot
(self, input_char, onehot_dim=38)
modules/prediction.py:16
↓ 2 callersMethod_conv3x3
3x3 convolution with padding
modules/feature_extraction.py:130
↓ 2 callersMethodadd
(self, v)
utils.py:156
↓ 2 callersFunctioncompute_partial_repr
(input_points, control_points)
modules/tps_spatial_transformer.py:20
↓ 2 callersMethodgenerate_1d_decay
generate 1d decay mask, the result is l*l
modules/VIPTRv1-T_ch.py:104
↓ 2 callersMethodgenerate_1d_decay
generate 1d decay mask, the result is l*l
modules/VIPTRv2T_ch.py:101
↓ 2 callersMethodgenerate_1d_decay
generate 1d decay mask, the result is l*l
modules/VIPTRv1.py:104
↓ 2 callersMethodgenerate_1d_decay
generate 1d decay mask, the result is l*l
modules/VIPTRv2.py:101
↓ 2 callersMethodim2cswin
(self, x, size)
modules/VIPTRv1-T_ch.py:604
↓ 2 callersMethodim2cswin
(self, x, size)
modules/VIPTRv2T_ch.py:533
↓ 2 callersMethodim2cswin
(self, x, size)
modules/VIPTRv1.py:604
↓ 2 callersMethodim2cswin
(self, x, size)
modules/VIPTRv2.py:592
↓ 2 callersMethodinit_weights
(self, module)
modules/stn_head.py:53
↓ 2 callersFunctionlr_warm
(base_lr, epoch, warm_epoch)
optimizer.py:219
↓ 2 callersMethodreset
(self)
utils.py:162
↓ 2 callersFunctiontruncated_normal_
(tensor, mean=0, std=0.02)
modules/SVTR.py:12
↓ 2 callersMethodval
(self)
utils.py:166
↓ 1 callersFunctionAdd_Padding
(image, top, bottom, left, right, color=(255,255,255))
dataload/loader.py:18
↓ 1 callersFunctionVIPTRv1
(opt)
modules/VIPTRv1-T_ch.py:987
↓ 1 callersFunctionVIPTRv1L
(opt)
modules/VIPTRv1.py:996
↓ 1 callersFunctionVIPTRv2
(opt)
modules/VIPTRv2.py:1038
↓ 1 callersFunctionVIPTRv2B
(opt)
modules/VIPTRv2.py:1057
↓ 1 callersMethod__init__
(self, input_size, hidden_size, num_classes)
modules/prediction.py:9
↓ 1 callersMethod_build_C
Return coordinates of fiducial points in I_r; C
modules/transformation.py:104
↓ 1 callersMethod_build_P
(self, I_r_width, I_r_height)
modules/transformation.py:136
↓ 1 callersMethod_build_P_hat
(self, F, C, P)
modules/transformation.py:145
↓ 1 callersMethod_build_inv_delta_C
Return inv_delta_C which is needed to calculate T
modules/transformation.py:114
↓ 1 callersMethod_forward_impl
(self, x, relative_pos_enc=None)
modules/VIPTRv1-T_ch.py:500
↓ 1 callersMethod_forward_impl
(self, x, relative_pos_enc=None)
modules/VIPTRv2T_ch.py:432
↓ 1 callersMethod_forward_impl
(self, x, relative_pos_enc=None)
modules/VIPTRv1.py:500
↓ 1 callersMethod_forward_impl
(self, x, relative_pos_enc=None)
modules/VIPTRv2.py:491
↓ 1 callersFunctionadd_gasuss_noise
Gasuss noise
dataload/dataAug.py:108
↓ 1 callersFunctionbenchmark_all_eval
evaluation with 6 benchmark evaluation datasets
test_benchmark.py:22
↓ 1 callersFunctionblur
blur
dataload/dataAug.py:81
↓ 1 callersMethodbuild_P_prime
Generate Grid from batch_C_prime [batch_size x F x 2]
modules/transformation.py:155
↓ 1 callersFunctionbuild_output_control_points
(num_control_points, margins)
modules/tps_spatial_transformer.py:36
↓ 1 callersMethodcalc_delta
(self)
dataload/aug/warp_mls.py:27
↓ 1 callersFunctioncvtColor
cvtColor
dataload/dataAug.py:70
↓ 1 callersFunctiondrop_path
Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks). the original name is misleading as 'Drop Connect' is a di
modules/SVTR.py:23
↓ 1 callersMethodencode
convert text-label into text-index. input: text: text labels of each image. [batch_size] batch_max_length: max length
utils.py:118
↓ 1 callersFunctionfixkeyCh
(key)
dataload/loader.py:24
↓ 1 callersFunctionfixkeyEn
(key)
dataload/loader.py:27
↓ 1 callersMethodforward_features
(self, x)
modules/SVTR.py:551
↓ 1 callersMethodforward_features
(self, x)
modules/VIPTRv1-T_ch.py:962
↓ 1 callersMethodforward_features
(self, x)
modules/VIPTRv2T_ch.py:841
↓ 1 callersMethodforward_features
(self, x)
modules/VIPTRv1.py:956
↓ 1 callersMethodforward_features
(self, x)
modules/VIPTRv2.py:1018
↓ 1 callersMethodgen_img
(self)
dataload/aug/warp_mls.py:105
↓ 1 callersMethodgenerate_2d_decay
generate 2d decay mask, the result is (HW)*(HW)
modules/VIPTRv1-T_ch.py:91
↓ 1 callersMethodgenerate_2d_decay
generate 2d decay mask, the result is (HW)*(HW)
modules/VIPTRv2T_ch.py:88
↓ 1 callersMethodgenerate_2d_decay
generate 2d decay mask, the result is (HW)*(HW)
modules/VIPTRv1.py:91
↓ 1 callersMethodgenerate_2d_decay
generate 2d decay mask, the result is (HW)*(HW)
modules/VIPTRv2.py:88
↓ 1 callersMethodget_batch
(self)
dataset.py:81
↓ 1 callersFunctionget_crop
random crop
dataload/dataAug.py:120
↓ 1 callersMethodget_lepe
(self, x, func, size)
modules/VIPTRv1-T_ch.py:619
↓ 1 callersMethodget_lepe
(self, x, func, size)
modules/VIPTRv2T_ch.py:548
↓ 1 callersMethodget_lepe
(self, x, func, size)
modules/VIPTRv1.py:619
↓ 1 callersMethodget_lepe
(self, x, func, size)
modules/VIPTRv2.py:607
↓ 1 callersFunctiongrid_sample
(input, grid, canvas = None)
modules/tps_spatial_transformer.py:8
↓ 1 callersFunctionimg2windows
img: B C H W
modules/VIPTRv1-T_ch.py:732
↓ 1 callersFunctionimg2windows
img: B C H W
modules/VIPTRv2T_ch.py:662
↓ 1 callersFunctionimg2windows
img: B C H W
modules/VIPTRv1.py:732
↓ 1 callersFunctionimg2windows
img: B C H W
modules/VIPTRv2.py:719
↓ 1 callersMethodinit_stn
(self, stn_fc2)
modules/stn_head.py:67
↓ 1 callersFunctionjitter
jitter
dataload/dataAug.py:92
↓ 1 callersMethodmake
make
dataload/dataAug.py:203
↓ 1 callersMethodmerge_ext_data
(self, data, ext_data)
dataload/dataAug.py:38
↓ 1 callersFunctionmotion_blur
(image)
dataload/dataAug.py:167
↓ 1 callersFunctionrandom_dilute
(image, sele_value=50, set_value=20, num_ratio=[0.1, 0.2])
dataload/dataAug.py:147
↓ 1 callersMethodreset_parameters
(self)
modules/VIPTRv1-T_ch.py:155
↓ 1 callersMethodreset_parameters
(self)
modules/VIPTRv2T_ch.py:151
↓ 1 callersMethodreset_parameters
(self)
modules/VIPTRv1.py:155
↓ 1 callersMethodreset_parameters
(self)
modules/VIPTRv2.py:151
↓ 1 callersFunctionresizeAug
(img, HThresh=25)
dataload/dataAug.py:138
↓ 1 callersFunctiontest
model configuration
test_benchmark.py:179
next →1–100 of 373, ranked by callers