MCPcopy Create free account

hub / github.com/TNTWEN/Pruned-OpenVINO-YOLO / functions

Functions172 in github.com/TNTWEN/Pruned-OpenVINO-YOLO

⨍ Functions 172◇ Types & classes 22
↓ 11 callersFunctiongather_bn_weights
(module_list, prune_idx)
Pruneyolov3v4/utils/prune_utils.py:129
↓ 9 callersFunctionload_darknet_weights
(self, weights, cutoff=-1)
Pruneyolov3v4/models.py:368
↓ 8 callersFunctionfitness
(k)
Pruneyolov3v4/utils/utils.py:685
↓ 8 callersFunctionobtain_bn_mask
(bn_module, thre)
Pruneyolov3v4/utils/prune_utils.py:312
↓ 7 callersFunctiontest
(cfg, data, weights=None, batch_size=16, imgsz=416, conf_thres=0.
Pruneyolov3v4/test.py:11
↓ 6 callersFunctioncheck_file
(file)
Pruneyolov3v4/utils/utils.py:46
↓ 6 callersFunctionprune_model_keep_size2
(model, prune_idx, CBL_idx, CBLidx2mask)
Pruneyolov3v4/utils/prune_utils.py:335
↓ 6 callersFunctionsave_weights
(self, path='model.weights', cutoff=-1)
Pruneyolov3v4/models.py:420
↓ 5 callersFunctionletterbox
(img, new_shape=(416, 416), color=(114, 114, 114), auto=True, scaleFill=False, scaleup=True)
Pruneyolov3v4/utils/datasets.py:621
↓ 5 callersFunctionupdate_activation
(i, pruned_model, activation, CBL_idx)
Pruneyolov3v4/utils/prune_utils.py:321
↓ 5 callersFunctionwrite_cfg
(cfg_file, module_defs)
Pruneyolov3v4/utils/prune_utils.py:142
↓ 5 callersFunctionxywh2xyxy
(x)
Pruneyolov3v4/utils/utils.py:113
↓ 5 callersFunctionxyxy2xywh
(x)
Pruneyolov3v4/utils/utils.py:103
↓ 4 callersFunctioncreate_folder
(path='./new_folder')
Pruneyolov3v4/utils/datasets.py:841
↓ 4 callersFunctioninit_weights_from_loose_model
(compact_model, loose_model, CBL_idx, Conv_idx, CBLidx2mask)
Pruneyolov3v4/utils/prune_utils.py:250
↓ 4 callersFunctionparse_module_defs2
(module_defs)
Pruneyolov3v4/utils/prune_utils.py:57
↓ 3 callersMethod__init__
(self, dimension=1)
Pruneyolov3v4/utils/layers.py:20
↓ 3 callersFunctionattempt_download
(weights)
Pruneyolov3v4/models.py:477
↓ 3 callersMethodcreate_grids
(self, ng=(13, 13), device='cpu')
Pruneyolov3v4/models.py:165
↓ 3 callersFunctionload_image
(self, index)
Pruneyolov3v4/utils/datasets.py:527
↓ 3 callersFunctionobtain_avg_forward_time
(input, model, repeat=200)
Pruneyolov3v4/layer_channel_prune.py:341
↓ 3 callersFunctionplot_images
(images, targets, paths=None, fname='images.jpg', names=None, max_size=640, max_subplots=16)
Pruneyolov3v4/utils/utils.py:859
↓ 3 callersFunctionprint_results
(k)
Pruneyolov3v4/utils/utils.py:673
↓ 3 callersFunctionscale_coords
(img1_shape, coords, img0_shape, ratio_pad=None)
Pruneyolov3v4/utils/utils.py:123
↓ 3 callersFunctionwh_iou
(wh1, wh2)
Pruneyolov3v4/utils/utils.py:310
↓ 2 callersMethodbackward
(ctx, grad_output)
Pruneyolov3v4/utils/layers.py:124
↓ 2 callersFunctionbox_area
(box)
Pruneyolov3v4/utils/utils.py:298
↓ 2 callersFunctionbox_iou
Return intersection-over-union (Jaccard index) of boxes. Both sets of boxes are expected to be in (x1, y1, x2, y2) format. Arguments:
Pruneyolov3v4/utils/utils.py:285
↓ 2 callersFunctionclip_coords
(boxes, img_shape)
Pruneyolov3v4/utils/utils.py:139
↓ 2 callersFunctioncompute_loss
(p, targets, model)
Pruneyolov3v4/utils/utils.py:353
↓ 2 callersMethodforward_once
(self, x, augment=False, verbose=False)
Pruneyolov3v4/models.py:293
↓ 2 callersMethodfuse
(self)
Pruneyolov3v4/models.py:343
↓ 2 callersFunctionget_input_mask
(module_defs, idx, CBLidx2mask)
Pruneyolov3v4/utils/prune_utils.py:197
↓ 2 callersFunctionhist2d
(x, y, n=100)
Pruneyolov3v4/utils/utils.py:990
↓ 2 callersMethodinfo
(self, verbose=False)
Pruneyolov3v4/models.py:360
↓ 2 callersFunctionload_classes
(path)
Pruneyolov3v4/utils/utils.py:56
↓ 2 callersFunctionmerge_mask
(model, CBLidx2mask, CBLidx2filters)
Pruneyolov3v4/utils/prune_utils.py:446
↓ 2 callersMethodnew_video
(self, path)
Pruneyolov3v4/utils/datasets.py:112
↓ 2 callersFunctionnon_max_suppression
Performs Non-Maximum Suppression on inference results Returns detections with shape: nx6 (x1, y1, x2, y2, conf, cls)
Pruneyolov3v4/utils/utils.py:479
↓ 2 callersFunctionobtain_avg_forward_time
(input, model, repeat=200)
Pruneyolov3v4/shortcut_prune.py:235
↓ 2 callersFunctionobtain_avg_forward_time
(input, model, repeat=200)
Pruneyolov3v4/layer_prune.py:220
↓ 2 callersFunctionobtain_avg_forward_time
(input, model, repeat=200)
Pruneyolov3v4/slim_prune.py:186
↓ 2 callersFunctionobtain_avg_forward_time
(input, model, repeat=200)
Pruneyolov3v4/prune.py:183
↓ 2 callersFunctionparse_data_cfg
(path)
Pruneyolov3v4/utils/parse_config.py:54
↓ 2 callersFunctionparse_module_defs
(module_defs)
Pruneyolov3v4/utils/prune_utils.py:13
↓ 2 callersFunctionparse_module_defs4
(module_defs)
Pruneyolov3v4/utils/prune_utils.py:107
↓ 2 callersFunctionplot_one_box
(x, img, color=None, label=None, line_thickness=None)
Pruneyolov3v4/utils/utils.py:825
↓ 2 callersFunctionrandom_affine
(img, targets=(), degrees=10, translate=.1, scale=.1, shear=10, border=0)
Pruneyolov3v4/utils/datasets.py:654
↓ 2 callersMethodstep
Performs a single optimization step. Arguments: closure (callable, optional): A closure that reevaluates the model
Pruneyolov3v4/utils/adabound.py:51
↓ 2 callersFunctiontrain
(hyp)
Pruneyolov3v4/train.py:65
↓ 1 callersMethod__init__
(self, cfg, img_size=(416, 416), verbose=False)
Pruneyolov3v4/models.py:238
↓ 1 callersMethod__init__
(self, params, lr=1e-3, betas=(0.9, 0.999), final_lr=0.1, gamma=1e-3, eps=1e-8, weight_decay=
Pruneyolov3v4/utils/adabound.py:26
↓ 1 callersMethod__setstate__
(self, state)
Pruneyolov3v4/utils/adabound.py:46
↓ 1 callersFunctionadjust_learning_rate
调整学习率进行warm up和学习率衰减
Pruneyolov3v4/train.py:173
↓ 1 callersFunctionap_per_class
Compute the average precision, given the recall and precision curves. Source: https://github.com/rafaelpadilla/Object-Detection-Metrics. # Ar
Pruneyolov3v4/utils/utils.py:147
↓ 1 callersFunctionapply_classifier
(x, model, img, im0)
Pruneyolov3v4/utils/utils.py:758
↓ 1 callersFunctionaugment_hsv
(img, hgain=0.5, sgain=0.5, vgain=0.5)
Pruneyolov3v4/utils/datasets.py:544
↓ 1 callersFunctionbbox_ioa
(box1, box2)
Pruneyolov3v4/utils/datasets.py:729
↓ 1 callersFunctionbbox_iou
(box1, box2, x1y1x2y2=True, GIoU=False, DIoU=False, CIoU=False)
Pruneyolov3v4/utils/utils.py:239
↓ 1 callersFunctionbuild_targets
(p, targets, model)
Pruneyolov3v4/utils/utils.py:420
↓ 1 callersFunctioncheck_git_status
()
Pruneyolov3v4/utils/utils.py:38
↓ 1 callersFunctioncoco80_to_coco91_class
()
Pruneyolov3v4/utils/utils.py:91
↓ 1 callersFunctioncompute_ap
Compute the average precision, given the recall and precision curves. Source: https://github.com/rbgirshick/py-faster-rcnn. # Arguments
Pruneyolov3v4/utils/utils.py:210
↓ 1 callersFunctioncreate_modules
(module_defs, img_size, cfg)
Pruneyolov3v4/models.py:8
↓ 1 callersFunctiondetect
(save_img=False)
Pruneyolov3v4/detect.py:8
↓ 1 callersFunctionexif_size
(img)
Pruneyolov3v4/utils/datasets.py:29
↓ 1 callersFunctiongdrive_download
(id='1HaXkef9z6y5l4vUnCYgdmEAj61c6bfWO', name='coco.zip')
Pruneyolov3v4/utils/google_utils.py:11
↓ 1 callersFunctionget_sr_flag
(epoch, sr)
Pruneyolov3v4/utils/prune_utils.py:8
↓ 1 callersFunctionget_yolo_layers
(model)
Pruneyolov3v4/models.py:364
↓ 1 callersFunctioninit_seeds
(seed=0)
Pruneyolov3v4/utils/utils.py:32
↓ 1 callersFunctionlabels_to_class_weights
(labels, nc=80)
Pruneyolov3v4/utils/utils.py:63
↓ 1 callersFunctionlabels_to_image_weights
(labels, nc=80, class_weights=np.ones(80))
Pruneyolov3v4/utils/utils.py:82
↓ 1 callersFunctionload_mosaic
(self, index)
Pruneyolov3v4/utils/datasets.py:563
↓ 1 callersFunctionobtain_filters_mask
(model, thre, CBL_idx, prune_idx)
Pruneyolov3v4/shortcut_prune.py:154
↓ 1 callersFunctionobtain_filters_mask
(model, CBL_idx, prune_shortcuts)
Pruneyolov3v4/layer_prune.py:118
↓ 1 callersFunctionobtain_filters_mask
(model, thre, CBL_idx, prune_idx)
Pruneyolov3v4/slim_prune.py:84
↓ 1 callersFunctionobtain_filters_mask
(model, thre, CBL_idx, prune_idx)
Pruneyolov3v4/prune.py:117
↓ 1 callersFunctionobtain_filters_mask
(model, thre, CBL_idx, prune_idx)
Pruneyolov3v4/layer_channel_prune.py:84
↓ 1 callersFunctionobtain_filters_mask2
(model, CBL_idx, prune_shortcuts)
Pruneyolov3v4/layer_channel_prune.py:239
↓ 1 callersFunctionoutput_to_target
Convert a YOLO model output to target format [batch_id, class_id, x, y, w, h, conf]
Pruneyolov3v4/utils/utils.py:799
↓ 1 callersFunctionparse_model_cfg
(path)
Pruneyolov3v4/utils/parse_config.py:6
↓ 1 callersFunctionplot_results
(start=0, stop=0, bucket='', id=())
Pruneyolov3v4/utils/utils.py:1051
↓ 1 callersFunctionprint_mutation
(hyp, results, bucket='')
Pruneyolov3v4/utils/utils.py:739
↓ 1 callersFunctionprune_and_eval
(model, sorted_bn, percent=.0)
Pruneyolov3v4/shortcut_prune.py:93
↓ 1 callersFunctionprune_and_eval
(model, prune_shortcuts=[])
Pruneyolov3v4/layer_prune.py:99
↓ 1 callersFunctionprune_and_eval
(model, CBL_idx, CBLidx2mask)
Pruneyolov3v4/slim_prune.py:140
↓ 1 callersFunctionprune_and_eval
(model, sorted_bn, percent=.0)
Pruneyolov3v4/prune.py:86
↓ 1 callersFunctionprune_and_eval
(model, CBL_idx, CBLidx2mask)
Pruneyolov3v4/layer_channel_prune.py:134
↓ 1 callersFunctionprune_and_eval2
(model, prune_shortcuts=[])
Pruneyolov3v4/layer_channel_prune.py:219
↓ 1 callersFunctionsmooth_BCE
(eps=0.1)
Pruneyolov3v4/utils/utils.py:348
↓ 1 callersFunctionstrip_optimizer
(f='weights/best.pt')
Pruneyolov3v4/utils/utils.py:579
↓ 1 callersMethodupdate
(self, model)
Pruneyolov3v4/utils/torch_utils.py:176
↓ 1 callersMethodupdateBN
(sr_flag, module_list, s, prune_idx, epoch, idx2mask=None, opt=None)
Pruneyolov3v4/utils/prune_utils.py:168
↓ 1 callersMethodupdate_attr
(self, model)
Pruneyolov3v4/utils/torch_utils.py:190
Method__getitem__
(self, index)
Pruneyolov3v4/utils/datasets.py:439
Method__init__
(self, anchors, nc, img_size, yolo_index, layers, stride)
Pruneyolov3v4/models.py:147
Method__init__
(self,layers,groupflag)
Pruneyolov3v4/utils/layers.py:29
Method__init__
(self, layers, weight=False)
Pruneyolov3v4/utils/layers.py:44
Method__init__
(self, in_ch, out_ch, k=(3, 5, 7), stride=1, dilation=1, bias=True, method='equal_params')
Pruneyolov3v4/utils/layers.py:76
Method__init__
(self, model, decay=0.9999, device='')
Pruneyolov3v4/utils/torch_utils.py:164
next →1–100 of 172, ranked by callers