keras-segnet's People
Contributors
Watchers
keras-segnet's Issues
Image dimensions incongruent after down and up sampling procedures
Problem
Sample Images have dimensions 11164 by 874
During the encoding portion of the neural net, the downsampling becomes inaccurate.
1164 / 2 = 582
582 / 2 = 291
291/ 2 = XXX
__________________________________________________________________________________________________
Layer (type) Output Shape Param # Connected to
==================================================================================================
input_1 (InputLayer) [(None, 1164, 874, 3 0
__________________________________________________________________________________________________
conv2d (Conv2D) (None, 582, 437, 32) 896 input_1[0][0]
__________________________________________________________________________________________________
batch_normalization (BatchNorma (None, 582, 437, 32) 128 conv2d[0][0]
__________________________________________________________________________________________________
activation (Activation) (None, 582, 437, 32) 0 batch_normalization[0][0]
__________________________________________________________________________________________________
activation_1 (Activation) (None, 582, 437, 32) 0 activation[0][0]
__________________________________________________________________________________________________
separable_conv2d (SeparableConv (None, 582, 437, 64) 2400 activation_1[0][0]
__________________________________________________________________________________________________
batch_normalization_1 (BatchNor (None, 582, 437, 64) 256 separable_conv2d[0][0]
__________________________________________________________________________________________________
activation_2 (Activation) (None, 582, 437, 64) 0 batch_normalization_1[0][0]
__________________________________________________________________________________________________
separable_conv2d_1 (SeparableCo (None, 582, 437, 64) 4736 activation_2[0][0]
__________________________________________________________________________________________________
batch_normalization_2 (BatchNor (None, 582, 437, 64) 256 separable_conv2d_1[0][0]
__________________________________________________________________________________________________
max_pooling2d (MaxPooling2D) (None, 291, 219, 64) 0 batch_normalization_2[0][0]
__________________________________________________________________________________________________
conv2d_1 (Conv2D) (None, 291, 219, 64) 2112 activation[0][0]
__________________________________________________________________________________________________
add (Add) (None, 291, 219, 64) 0 max_pooling2d[0][0]
conv2d_1[0][0]
__________________________________________________________________________________________________
activation_3 (Activation) (None, 291, 219, 64) 0 add[0][0]
__________________________________________________________________________________________________
separable_conv2d_2 (SeparableCo (None, 291, 219, 128 8896 activation_3[0][0]
__________________________________________________________________________________________________
batch_normalization_3 (BatchNor (None, 291, 219, 128 512 separable_conv2d_2[0][0]
__________________________________________________________________________________________________
activation_4 (Activation) (None, 291, 219, 128 0 batch_normalization_3[0][0]
__________________________________________________________________________________________________
separable_conv2d_3 (SeparableCo (None, 291, 219, 128 17664 activation_4[0][0]
__________________________________________________________________________________________________
batch_normalization_4 (BatchNor (None, 291, 219, 128 512 separable_conv2d_3[0][0]
__________________________________________________________________________________________________
max_pooling2d_1 (MaxPooling2D) (None, 146, 110, 128 0 batch_normalization_4[0][0]
__________________________________________________________________________________________________
conv2d_2 (Conv2D) (None, 146, 110, 128 8320 add[0][0]
__________________________________________________________________________________________________
add_1 (Add) (None, 146, 110, 128 0 max_pooling2d_1[0][0]
conv2d_2[0][0]
__________________________________________________________________________________________________
activation_5 (Activation) (None, 146, 110, 128 0 add_1[0][0]
__________________________________________________________________________________________________
separable_conv2d_4 (SeparableCo (None, 146, 110, 256 34176 activation_5[0][0]
__________________________________________________________________________________________________
batch_normalization_5 (BatchNor (None, 146, 110, 256 1024 separable_conv2d_4[0][0]
__________________________________________________________________________________________________
activation_6 (Activation) (None, 146, 110, 256 0 batch_normalization_5[0][0]
__________________________________________________________________________________________________
separable_conv2d_5 (SeparableCo (None, 146, 110, 256 68096 activation_6[0][0]
__________________________________________________________________________________________________
batch_normalization_6 (BatchNor (None, 146, 110, 256 1024 separable_conv2d_5[0][0]
__________________________________________________________________________________________________
max_pooling2d_2 (MaxPooling2D) (None, 73, 55, 256) 0 batch_normalization_6[0][0]
__________________________________________________________________________________________________
conv2d_3 (Conv2D) (None, 73, 55, 256) 33024 add_1[0][0]
__________________________________________________________________________________________________
add_2 (Add) (None, 73, 55, 256) 0 max_pooling2d_2[0][0]
conv2d_3[0][0]
__________________________________________________________________________________________________
activation_7 (Activation) (None, 73, 55, 256) 0 add_2[0][0]
__________________________________________________________________________________________________
conv2d_transpose (Conv2DTranspo (None, 73, 55, 256) 590080 activation_7[0][0]
__________________________________________________________________________________________________
batch_normalization_7 (BatchNor (None, 73, 55, 256) 1024 conv2d_transpose[0][0]
__________________________________________________________________________________________________
activation_8 (Activation) (None, 73, 55, 256) 0 batch_normalization_7[0][0]
__________________________________________________________________________________________________
conv2d_transpose_1 (Conv2DTrans (None, 73, 55, 256) 590080 activation_8[0][0]
__________________________________________________________________________________________________
batch_normalization_8 (BatchNor (None, 73, 55, 256) 1024 conv2d_transpose_1[0][0]
__________________________________________________________________________________________________
up_sampling2d_1 (UpSampling2D) (None, 146, 110, 256 0 add_2[0][0]
__________________________________________________________________________________________________
up_sampling2d (UpSampling2D) (None, 146, 110, 256 0 batch_normalization_8[0][0]
__________________________________________________________________________________________________
conv2d_4 (Conv2D) (None, 146, 110, 256 65792 up_sampling2d_1[0][0]
__________________________________________________________________________________________________
add_3 (Add) (None, 146, 110, 256 0 up_sampling2d[0][0]
conv2d_4[0][0]
__________________________________________________________________________________________________
activation_9 (Activation) (None, 146, 110, 256 0 add_3[0][0]
__________________________________________________________________________________________________
conv2d_transpose_2 (Conv2DTrans (None, 146, 110, 128 295040 activation_9[0][0]
__________________________________________________________________________________________________
batch_normalization_9 (BatchNor (None, 146, 110, 128 512 conv2d_transpose_2[0][0]
__________________________________________________________________________________________________
activation_10 (Activation) (None, 146, 110, 128 0 batch_normalization_9[0][0]
__________________________________________________________________________________________________
conv2d_transpose_3 (Conv2DTrans (None, 146, 110, 128 147584 activation_10[0][0]
__________________________________________________________________________________________________
batch_normalization_10 (BatchNo (None, 146, 110, 128 512 conv2d_transpose_3[0][0]
__________________________________________________________________________________________________
up_sampling2d_3 (UpSampling2D) (None, 292, 220, 256 0 add_3[0][0]
__________________________________________________________________________________________________
up_sampling2d_2 (UpSampling2D) (None, 292, 220, 128 0 batch_normalization_10[0][0]
__________________________________________________________________________________________________
conv2d_5 (Conv2D) (None, 292, 220, 128 32896 up_sampling2d_3[0][0]
__________________________________________________________________________________________________
add_4 (Add) (None, 292, 220, 128 0 up_sampling2d_2[0][0]
conv2d_5[0][0]
__________________________________________________________________________________________________
activation_11 (Activation) (None, 292, 220, 128 0 add_4[0][0]
__________________________________________________________________________________________________
conv2d_transpose_4 (Conv2DTrans (None, 292, 220, 64) 73792 activation_11[0][0]
__________________________________________________________________________________________________
batch_normalization_11 (BatchNo (None, 292, 220, 64) 256 conv2d_transpose_4[0][0]
__________________________________________________________________________________________________
activation_12 (Activation) (None, 292, 220, 64) 0 batch_normalization_11[0][0]
__________________________________________________________________________________________________
conv2d_transpose_5 (Conv2DTrans (None, 292, 220, 64) 36928 activation_12[0][0]
__________________________________________________________________________________________________
batch_normalization_12 (BatchNo (None, 292, 220, 64) 256 conv2d_transpose_5[0][0]
__________________________________________________________________________________________________
up_sampling2d_5 (UpSampling2D) (None, 584, 440, 128 0 add_4[0][0]
__________________________________________________________________________________________________
up_sampling2d_4 (UpSampling2D) (None, 584, 440, 64) 0 batch_normalization_12[0][0]
__________________________________________________________________________________________________
conv2d_6 (Conv2D) (None, 584, 440, 64) 8256 up_sampling2d_5[0][0]
__________________________________________________________________________________________________
add_5 (Add) (None, 584, 440, 64) 0 up_sampling2d_4[0][0]
conv2d_6[0][0]
__________________________________________________________________________________________________
activation_13 (Activation) (None, 584, 440, 64) 0 add_5[0][0]
__________________________________________________________________________________________________
conv2d_transpose_6 (Conv2DTrans (None, 584, 440, 32) 18464 activation_13[0][0]
__________________________________________________________________________________________________
batch_normalization_13 (BatchNo (None, 584, 440, 32) 128 conv2d_transpose_6[0][0]
__________________________________________________________________________________________________
activation_14 (Activation) (None, 584, 440, 32) 0 batch_normalization_13[0][0]
__________________________________________________________________________________________________
conv2d_transpose_7 (Conv2DTrans (None, 584, 440, 32) 9248 activation_14[0][0]
__________________________________________________________________________________________________
batch_normalization_14 (BatchNo (None, 584, 440, 32) 128 conv2d_transpose_7[0][0]
__________________________________________________________________________________________________
up_sampling2d_7 (UpSampling2D) (None, 1168, 880, 64 0 add_5[0][0]
__________________________________________________________________________________________________
up_sampling2d_6 (UpSampling2D) (None, 1168, 880, 32 0 batch_normalization_14[0][0]
__________________________________________________________________________________________________
conv2d_7 (Conv2D) (None, 1168, 880, 32 2080 up_sampling2d_7[0][0]
__________________________________________________________________________________________________
add_6 (Add) (None, 1168, 880, 32 0 up_sampling2d_6[0][0]
conv2d_7[0][0]
__________________________________________________________________________________________________
conv2d_8 (Conv2D) (None, 1168, 880, 6) 1734 add_6[0][0]
==================================================================================================
Total params: 2,059,846
Trainable params: 2,056,070
Non-trainable params: 3,776
__________________________________________________________________________________________________
You can see when the output shape is upsampled it comes out to 1168 causing an inconsistency.
Error:
logits and labels must have the same first dimension, got logits shape [2055680,6] and labels shape [6104016]
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.