Functions — github.com/BBuf/Keras-Semantic-Segmentation

hub / github.com/BBuf/Keras-Semantic-Segmentation / functions

Functions163 in github.com/BBuf/Keras-Semantic-Segmentation

⨍ Functions 163◇ Types & classes 4

↓ 1 callersFunctionmain

()

json_to_dataset.py:22

↓ 1 callersFunctionmask_aug

()

albumentations.py:29

↓ 1 callersFunctionpool_block

(inp, pool_factor)

Models/PSPNet.py:8

↓ 1 callersFunctionres_xception_block

(x, channels)

Models/DeepLabV3+.py:97

↓ 1 callersFunctionres_xception_downsample_block

(x, channels)

Models/DeepLabV3+.py:65

↓ 1 callersFunctionsse_block

(prevlayer, prefix)

Models/scSEUnet.py:21

↓ 1 callersFunctionxception_block

(x, channels)

Models/DeepLabV3+.py:73

FunctionAttUNet

(nClasses, input_height=224, input_width=224)

Models/AttUNet.py:35

FunctionBasicBlock

BasicBlock搭建 Args: x: 输入张量 out_channels: 输出通道数 stride: 步长 downsample: 是否下采样 Returns:

Models/ACNet.py:87

FunctionBottleNeckBlock

BottleNeckBlock搭建 Args: x: 输入张量 out_channels: 输出通道数 stride: 步长 downsample: 是否下采样 Returns:

Models/ACNet.py:114

FunctionDeepLabV2

(nClasses, input_height=224, input_width=224)

Models/DeepLabV2.py:20

FunctionDeeplabV3_plus

(nClasses=21, input_height=512, input_width=512, out_stride=16)

Models/DeepLabV3+.py:141

FunctionENet

(n_classes, input_height=256, input_width=256)

Models/ENet.py:133

FunctionFCN8

(nClasses, input_height=224, input_width=224)

Models/FCN8.py:54

FunctionGeneralized_Dice_Loss

(y_true, y_pred)

losses/Generalized_Dice_loss.py:19

FunctionGhostNet

(x, num_classes=1000, width_mult=1.)

Models/GhostNet.py:87

FunctionHRNet

(nClasses, input_height=224, input_width=224)

Models/HRNet.py:146

FunctionICNet

(nClasses, input_height=224, input_width=224)

Models/ICNet.py:7

FunctionMobileNetFCN8

(nClasses, optimizer=None, input_width=512, input_height=512, pretrained='imagenet')

Models/MobileNetFCN8.py:29

FunctionMobileNetUnet

(nClasses, input_width=224, input_height=224)

Models/MobileNetUnet.py:27

FunctionMobileNext

MobileNext网络主体 Args: x: 输入张量 num_classes: 最后输出分类数目 width_mult: 宽度扩张 identity_tensor_multiplier: 参见SandGlass模块

Models/MobileNext.py:138

FunctionNestedUNet

(nClasses, input_height=224, input_width=224)

Models/NestedUNet.py:17

FunctionPSPNet

(nClasses, input_width=384, input_height=384)

Models/PSPNet.py:19

FunctionR2AttUNet

(nClasses, input_height=224, input_width=224)

Models/R2AttUNet.py:55

FunctionR2UNet

R2U-Unet implementation Paper: https://arxiv.org/abs/1802.06955

Models/R2UNet.py:41

FunctionResACNet

返回各个系列的由ACBlock组成的resnet Args: x: 输入张量 class_dim: 输出类别数 depth: 网络深度，参加下面assert允许的resnet深度 Returns: Usage：

Models/ACNet.py:174

FunctionSEUnet

(nClasses, input_height=224, input_width=224)

Models/SEUNet.py:18

FunctionSegnet

(nClasses, input_height=224, input_width=224)

Models/Segnet.py:5

FunctionUNet

UNet - Basic Implementation Paper : https://arxiv.org/abs/1505.04597

Models/Unet.py:23

FunctionUnet_Xception_ResNetBlock

(nClasses, input_height=224, input_width=224)

Models/UNet_Xception_ResNetBlock.py:24

FunctionVGGFCN8

(nClasses, input_height=224, input_width=224, vgg_weight_path=None)

Models/VGGFCN8.py:7

FunctionVGGUnet

(nClasses, input_height=224, input_width=224, vgg_weight_path=None)

Models/VGGUnet.py:7

Function_BCE_Dice_Loss

(y_true, y_pred)

losses/BCE_Dice_loss.py:7

Function_BCE_Jaccard_Loss

Sum of binary crossentropy and jaccard losses: .. math:: L(A, B) = bce_weight * binary_crossentropy(A, B) + jaccard_loss(A, B) A

losses/BCE_Jaccard_Loss.py:7

Function_CE_Dice_loss

(y_true, y_pred)

losses/CE_Dice_loss.py:20

Function_CE_Jaccard_Loss

Sum of categorical crossentropy and jaccard losses: .. math:: L(A, B) = cce_weight * categorical_crossentropy(A, B) + jaccard_loss(A, B)

losses/CE_Jaccard_Loss.py:6

Function_Focal_Tversky_Loss

(y_true, y_pred)

losses/Focal_Tversky_loss.py:16

Function_Jaccard_Loss

Jaccard loss function for imbalanced datasets: .. math:: L(A, B) = 1 - \frac{A \cap B}{A \cup B} Args: gt: ground truth 4D keras te

losses/Jaccard_loss.py:43

Function_Tversky_Loss

(y_true, y_pred)

losses/Tversky_loss.py:16

Method__init__

(self)

tools/keras2Msnh/MsnhBuilder.py:2

Method__init__

(self,model,filepath, monitor='val_iou_score', save_best_only=True, save_weights_only=True,

utils/utils.py:41

Method__init__

(self, upsampling=(2, 2), data_format=None, **kwargs)

Models/DeepLabV3+.py:16

Method__init__

(self, upsampling, **kwargs)

Models/DeepLabV2.py:8

Methodbuild

(self, input_shape)

Models/DeepLabV2.py:12

Methodcall

(self, inputs)

Models/DeepLabV3+.py:29

Methodcall

(self, x, mask=None)

Models/DeepLabV2.py:15

Methodcompute_output_shape

(self, input_shape)

Models/DeepLabV3+.py:22

(keras_model, caffe_net_file, caffe_params_file)

tools/keras2caffe/convert.py:42

Functionfocal_loss_fixed

(y_true, y_pred)

losses/B_Focal_loss.py:6

Functionfocal_loss_fixed

y_pred 是输出Tensor，形状类似[None, 10]，其中10是类别数 y_true 是标签Tensor

losses/C_Focal_loss.py:13

Functionget_iou_score

(class_weights=1., smooth=SMOOTH, per_image=True, threshold=None)

metrics/metrics.py:48

FunctionimageSegmentationGenerator

(images_path, segs_path, batch_size, n_classes, input_height, input_width, resi

data.py:45

Functioniou_score

参数： gt: ground truth 4D keras tensor (B, H, W, C) pr: prediction 4D keras tensor (B, H, W, C) class_weights: 1. or li

losses/Jaccard_loss.py:5

Functionloss_

(y_true, y_pred)

losses/Weighted_Categorical_loss.py:16

Functionmain

tools/keras2Msnh/project/unet.cpp:7

Models/MobileNetFCN8.py:7

Functionrelu6

(x)

Models/MobileNetUnet.py:6

FunctionscSEUnet

(nClasses, input_height=224, input_width=224)

Models/scSEUnet.py:46

Functionscore

(gt, pr)

metrics/metrics.py:49

Methodset_model

(self, model)

utils/utils.py:47

Functionshow_avai_models

Displays available models. Examples:: >>> from Models import models >>> models.show_avai_models()

Models/__init__.py:47

← previous101–163 of 163, ranked by callers