MCPcopy Create free account

hub / github.com/aojunzz/NM-sparsity / functions

Functions200 in github.com/aojunzz/NM-sparsity

↓ 10 callersMethodupdate
(self, val, n=1)
classification/train_imagenet.py:290
↓ 8 callersMethodstate_dict
Returns the state of the scheduler as a :class:`dict`. It contains an entry for every variable in self.__dict__ which is not the opti
RAFT/core/lr_scheduler.py:69
↓ 7 callersMethodload_state_dict
Loads the schedulers state. Args: state_dict (dict): scheduler state. Should be an object returned from a call to
RAFT/core/lr_scheduler.py:77
↓ 6 callersMethod__init__
(self, hidden_dim=128, input_dim=192+128)
RAFT/core/sparse_update.py:21
↓ 6 callersMethod__init__
(self, hidden_dim=128, input_dim=192+128)
RAFT/core/update.py:17
↓ 5 callersMethodpad
(self, *inputs)
RAFT/core/utils/utils.py:18
↓ 5 callersMethodupdate
(self)
RAFT/train.py:42
↓ 4 callersMethod__init__
(self, aug_params=None, root='datasets/HD1k')
RAFT/core/datasets.py:181
↓ 4 callersMethod_make_layer
(self, block, planes, blocks, stride=1, N = 2, M = 4)
classification/models/resnet.py:125
↓ 4 callersFunctioncoords_grid
(batch, ht, wd, device)
RAFT/core/utils/utils.py:74
↓ 4 callersMethodunpad
(self,x)
RAFT/core/utils/utils.py:21
↓ 3 callersMethod__init__
(self, output_dim=128, norm_fn='batch', dropout=0.0)
RAFT/core/extractor.py:119
↓ 3 callersMethod__init__
(self, output_dim=128, norm_fn='batch', dropout=0.0)
RAFT/core/sparse_extractor.py:124
↓ 3 callersMethod_format_param
Return correctly formatted lr/momentum for each param group.
RAFT/core/lr_scheduler.py:379
↓ 3 callersMethod_make_layer
(self, dim, stride=1)
RAFT/core/extractor.py:159
↓ 3 callersMethod_make_layer
(self, dim, stride=1)
RAFT/core/extractor.py:235
↓ 3 callersMethod_make_layer
(self, dim, stride=1)
RAFT/core/sparse_extractor.py:164
↓ 3 callersMethod_make_layer
(self, dim, stride=1)
RAFT/core/sparse_extractor.py:240
↓ 3 callersMethodstep
(self, optimizer)
RAFT/train.py:40
↓ 2 callersMethod__init__
(self, block, layers, num_classes=1000, N=2, M=4)
classification/models/resnet.py:100
↓ 2 callersFunctionaccuracy
Computes the precision@k for the specified values of k
classification/train_imagenet.py:296
↓ 2 callersMethodbackward
(ctx, grad_output)
devkit/sparse_ops/sparse_ops.py:32
↓ 2 callersMethodclose
(self)
RAFT/train.py:137
↓ 2 callersFunctionconv3x3
3x3 convolution with padding
classification/models/resnet.py:21
↓ 2 callersMethodfreeze_bn
(self)
RAFT/core/raft.py:58
↓ 2 callersMethodget_lr
(self)
RAFT/core/lr_scheduler.py:91
↓ 2 callersFunctionload_image
(imfile)
RAFT/demo.py:20
↓ 2 callersFunctionupflow8
(flow, mode='bilinear')
RAFT/core/utils/utils.py:80
↓ 2 callersFunctionvalidate
(val_loader, model, criterion, epoch, writer)
classification/train_imagenet.py:218
↓ 1 callersMethod__init__
(self, in_channels, out_channels, kernel_size, stride=1, padding=0, dilation=1, groups=1, bias=True, padding_m
devkit/sparse_ops/sparse_ops.py:68
↓ 1 callersMethod__init__
(self, enabled)
RAFT/core/sparse_raft.py:16
↓ 1 callersMethod_print_training_status
(self)
RAFT/train.py:102
↓ 1 callersFunctionaverage_gradients
(model)
devkit/core/dist_utils.py:23
↓ 1 callersFunctionbilinear_sampler
Wrapper for grid_sample, uses pixel coordinates
RAFT/core/utils/utils.py:57
↓ 1 callersFunctionbroadcast_params
(model)
devkit/core/dist_utils.py:28
↓ 1 callersMethodcolor_transform
Photometric augmentation
RAFT/core/utils/augmentor.py:36
↓ 1 callersMethodcolor_transform
(self, img1, img2)
RAFT/core/utils/augmentor.py:142
↓ 1 callersMethodcorr
(fmap1, fmap2)
RAFT/core/corr.py:53
↓ 1 callersFunctioncount_parameters
(model)
RAFT/train.py:80
↓ 1 callersFunctiondemo
(args)
RAFT/demo.py:42
↓ 1 callersMethoderaser_transform
Occlusion augmentation
RAFT/core/utils/augmentor.py:52
↓ 1 callersMethoderaser_transform
(self, img1, img2)
RAFT/core/utils/augmentor.py:148
↓ 1 callersFunctionfetch_optimizer
Create the optimizer and learning rate scheduler
RAFT/train.py:84
↓ 1 callersFunctionflow_uv_to_colors
Applies the flow color wheel to (possibly clipped) flow components u and v. According to the C++ source code of Daniel Scharstein Accord
RAFT/core/utils/flow_viz.py:70
↓ 1 callersMethodforward
Estimate optical flow between pair of frames
RAFT/core/raft.py:86
↓ 1 callersFunctionforward_interpolate
(flow)
RAFT/core/utils/utils.py:26
↓ 1 callersMethodget_last_lr
Return last computed learning rate by current scheduler.
RAFT/core/lr_scheduler.py:86
↓ 1 callersMethodget_sparse_weights
(self)
devkit/sparse_ops/sparse_ops.py:74
↓ 1 callersMethodget_sparse_weights
(self)
devkit/sparse_ops/sparse_ops.py:98
↓ 1 callersFunctioninit_dist
(backend='nccl', master_ip='127.0.0.1', port=29500)
devkit/core/dist_utils.py:9
↓ 1 callersMethodinitialize_flow
Flow is represented as difference between two coordinate grids flow = coords1 - coords0
RAFT/core/sparse_raft.py:63
↓ 1 callersMethodinitialize_flow
Flow is represented as difference between two coordinate grids flow = coords1 - coords0
RAFT/core/raft.py:63
↓ 1 callersFunctionload_state
(model_dir, model, optimizer=None)
devkit/core/utils.py:18
↓ 1 callersFunctionload_state_ckpt
(model_path, model)
devkit/core/utils.py:59
↓ 1 callersFunctionmain
()
classification/train_imagenet.py:40
↓ 1 callersFunctionmake_colorwheel
Generates a color wheel for optical flow visualization as presented in: Baker et al. "A Database and Evaluation Methodology for Optical F
RAFT/core/utils/flow_viz.py:20
↓ 1 callersFunctionpil_loader
(filename)
devkit/dataset/imagenet_dataset.py:5
↓ 1 callersMethodprint_lr
Display the current learning rate.
RAFT/core/lr_scheduler.py:95
↓ 1 callersMethodpush
(self, metrics)
RAFT/train.py:117
↓ 1 callersFunctionreadFlow
Read .flo file in Middlebury format
RAFT/core/utils/frame_utils.py:12
↓ 1 callersFunctionreadPFM
(file)
RAFT/core/utils/frame_utils.py:33
↓ 1 callersMethodreset
(self)
classification/train_imagenet.py:284
↓ 1 callersMethodreset_parameters
(self)
devkit/sparse_ops/syncbn_layer.py:83
↓ 1 callersMethodresize_sparse_flow_map
(self, flow, valid, fx=1.0, fy=1.0)
RAFT/core/utils/augmentor.py:161
↓ 1 callersFunctionsave_checkpoint
(model_dir, state, is_best)
devkit/core/utils.py:6
↓ 1 callersMethodscale
(self, loss)
RAFT/train.py:36
↓ 1 callersFunctionsequence_loss
Loss function defined over sequence of flow predictions
RAFT/train.py:52
↓ 1 callersMethodspatial_transform
(self, img1, img2, flow)
RAFT/core/utils/augmentor.py:67
↓ 1 callersMethodspatial_transform
(self, img1, img2, flow, valid)
RAFT/core/utils/augmentor.py:195
↓ 1 callersMethodstep
(self, epoch=None)
RAFT/core/lr_scheduler.py:107
↓ 1 callersFunctiontrain
(args)
RAFT/train.py:141
↓ 1 callersFunctiontrain
(train_loader, model, criterion, optimizer, lr_scheduler, epoch, writer)
classification/train_imagenet.py:152
↓ 1 callersMethodunscale_
(self, optimizer)
RAFT/train.py:38
↓ 1 callersMethodupsample_flow
Upsample flow field [H/8, W/8, 2] -> [H, W, 2] using convex combination
RAFT/core/sparse_raft.py:72
↓ 1 callersMethodupsample_flow
Upsample flow field [H/8, W/8, 2] -> [H, W, 2] using convex combination
RAFT/core/raft.py:72
↓ 1 callersFunctionvalidate_chairs
Perform evaluation on the FlyingChairs (test) split
RAFT/evaluate.py:75
↓ 1 callersFunctionvalidate_kitti
Peform validation using the KITTI-2015 (train) split
RAFT/evaluate.py:131
↓ 1 callersFunctionvalidate_sintel
Peform validation using the Sintel (train) split
RAFT/evaluate.py:96
↓ 1 callersFunctionviz
(img, flo)
RAFT/demo.py:26
FunctionPYBIND11_MODULE
RAFT/alt_cuda_corr/correlation.cpp:51
Method__call__
(self, tensor)
devkit/dataset/imagenet_dataset.py:52
Method__call__
(self, coords)
RAFT/core/corr.py:29
Method__call__
(self, coords)
RAFT/core/corr.py:74
Method__call__
(self, img1, img2, flow)
RAFT/core/utils/augmentor.py:111
Method__call__
(self, img1, img2, flow, valid)
RAFT/core/utils/augmentor.py:236
Method__enter__
(self)
RAFT/core/sparse_raft.py:18
Method__enter__
(self)
RAFT/core/lr_scheduler.py:132
Method__enter__
(self)
RAFT/core/raft.py:18
Method__exit__
(self, *args)
RAFT/core/sparse_raft.py:20
Method__exit__
(self, type, value, traceback)
RAFT/core/lr_scheduler.py:136
Method__exit__
(self, *args)
RAFT/core/raft.py:20
Method__getitem__
(self, idx)
devkit/dataset/imagenet_dataset.py:27
Method__getitem__
(self, index)
RAFT/core/datasets.py:34
Method__init__
(self, root_dir, meta_file, transform=None)
devkit/dataset/imagenet_dataset.py:11
Method__init__
(self, eig_vec=None, eig_val=None)
devkit/dataset/imagenet_dataset.py:40
Method__init__
(self, optimizer, niters, args)
devkit/core/lr_scheduler.py:44
Method__init__
(self, in_features: int, out_features: int, bias: bool = True, N=2, M=2, decay = 0.0002, **kwargs)
devkit/sparse_ops/sparse_ops.py:92
Method__init__
(self, num_features, eps=1e-5, momentum=0.9,last_gamma=False)
devkit/sparse_ops/syncbn_layer.py:68
Method__init__
(self, enabled=None)
RAFT/train.py:34
Method__init__
(self, model, scheduler)
RAFT/train.py:95
next →1–100 of 200, ranked by callers