Trjtdtk

Two field separators (colon and space) in awk

Can I criticise the more senior developers around me for not writing clean code?

What is the term for a person whose job is to place products on shelves in stores?

Extension of 2-adic valuation to the real numbers

"The cow" OR "a cow" OR "cows" in this context

Is there a way to generate a list of distinct numbers such that no two subsets ever have an equal sum?

Elements that can bond to themselves?

Determine the application client is using to connect

Pulling the rope with one hand is as heavy as with two hands?

Do I have an "anti-research" personality?

Usage of 万 (wàn): ten thousands or just a great number?

How does Nebula have access to these memories?

Critique of timeline aesthetic

Is there any official lore on the Far Realm?

How do I check if a string is entirely made of the same substring?

How to limit Drive Letters Windows assigns to new removable USB drives

Thesis on avalanche prediction using One Class SVM

A ​Note ​on ​N!

Read line from file and process something

Covering null sets by a finite number of intervals

Philosophical question on logistic regression: why isn't the optimal threshold value trained?

Retract an already submitted recommendation letter (written for an undergrad student)

What happens in the secondary winding if there's no spark plug connected?

If a planet has 3 moons, is it possible to have triple Full/New Moons at once?

Some question about output shape of joint coordinate in cnn-regression

feature extraction for a pretrained model in kerasMultiple Output Layers in Neural Networks in Deep Q LearningKeras input dimension bug?How to make output dimensions match input dimensions in CNN?Reshape output of convolutional layer to which dimensions?Keras CNN image input and outputUsing small CNN for de-noising on a full imageComparing tensors in TensorFlowInput shape for simpler time series in LSTM+CNNMultiple-input multiple-output CNN with custom loss function

1

$begingroup$

I have some question about output shape of cnn-regression problem.

Let say I have Image shape with

image shape = (105, 157, 3)

and this image has 14 joint coordinates for each image (14, 3, 10000).

I guess 10000 is the number of pictures

and each joint coordinates looks like

y=array([[145.82985678, 95.5022898 , 1. ],
 [111.87785389, 83.38437237, 1. ],
 [ 0. , -27.29260682, 0. ],
 [146.22287032, 67.62016267, 1. ],
 [114.716285 , 89.0394005 , 1. ],
 [147.03073148, 90.65512282, 1. ],
 [ 0. , -27.29260682, 0. ],
 [ 0. , -27.29260682, 0. ],
 [ 0. , -27.29260682, 0. ],
 [102.99138111, 63.60269094, 1. ],
 [107.44553454, 91.85599752, 1. ],
 [112.68571505, 118.12240233, 1. ],
 [ 95.72063065, 64.38871802, 1. ],
 [ 76.74681039, 60.37124629, 1. ]])

If I want to make a neural network then the neural network should have output and output should compare with this y true value

1. What is the output should shape with?
   




2. Which shape do I have to reshape this y true value? and how?
   

---

Dataset is from Leeds Sports Dataset (LSP): 11000 training and 1000 testing images from sports activities with challenging in terms of appearance and especially articulations. The majority of people have 150 pixel height. For each person the full body is labeled with total 14 joints.

python deep-learning keras cnn

share|improve this question

asked Apr 6 at 12:47

kim Yumikim Yumi

62

$endgroup$

• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  I guess the third coordinate is somehow a label rather than a coordinate location.
  $endgroup$
  – Vaalizaadeh
  Apr 6 at 13:06
  

  
  
  
  

add a comment | 

1

$begingroup$

I have some question about output shape of cnn-regression problem.

Let say I have Image shape with

image shape = (105, 157, 3)

and this image has 14 joint coordinates for each image (14, 3, 10000).

I guess 10000 is the number of pictures

and each joint coordinates looks like

y=array([[145.82985678, 95.5022898 , 1. ],
 [111.87785389, 83.38437237, 1. ],
 [ 0. , -27.29260682, 0. ],
 [146.22287032, 67.62016267, 1. ],
 [114.716285 , 89.0394005 , 1. ],
 [147.03073148, 90.65512282, 1. ],
 [ 0. , -27.29260682, 0. ],
 [ 0. , -27.29260682, 0. ],
 [ 0. , -27.29260682, 0. ],
 [102.99138111, 63.60269094, 1. ],
 [107.44553454, 91.85599752, 1. ],
 [112.68571505, 118.12240233, 1. ],
 [ 95.72063065, 64.38871802, 1. ],
 [ 76.74681039, 60.37124629, 1. ]])

If I want to make a neural network then the neural network should have output and output should compare with this y true value

1. What is the output should shape with?
   




2. Which shape do I have to reshape this y true value? and how?
   

---

Dataset is from Leeds Sports Dataset (LSP): 11000 training and 1000 testing images from sports activities with challenging in terms of appearance and especially articulations. The majority of people have 150 pixel height. For each person the full body is labeled with total 14 joints.

python deep-learning keras cnn

share|improve this question

asked Apr 6 at 12:47

kim Yumikim Yumi

62

$endgroup$

• 
  
  
  

  
  

  
  
  
  
  
  

  
  $begingroup$
  I guess the third coordinate is somehow a label rather than a coordinate location.
  $endgroup$
  – Vaalizaadeh
  Apr 6 at 13:06
  

  
  
  
  

add a comment | 

$begingroup$

I have some question about output shape of cnn-regression problem.

Let say I have Image shape with

image shape = (105, 157, 3)

and this image has 14 joint coordinates for each image (14, 3, 10000).

I guess 10000 is the number of pictures

and each joint coordinates looks like

y=array([[145.82985678, 95.5022898 , 1. ],
 [111.87785389, 83.38437237, 1. ],
 [ 0. , -27.29260682, 0. ],
 [146.22287032, 67.62016267, 1. ],
 [114.716285 , 89.0394005 , 1. ],
 [147.03073148, 90.65512282, 1. ],
 [ 0. , -27.29260682, 0. ],
 [ 0. , -27.29260682, 0. ],
 [ 0. , -27.29260682, 0. ],
 [102.99138111, 63.60269094, 1. ],
 [107.44553454, 91.85599752, 1. ],
 [112.68571505, 118.12240233, 1. ],
 [ 95.72063065, 64.38871802, 1. ],
 [ 76.74681039, 60.37124629, 1. ]])

If I want to make a neural network then the neural network should have output and output should compare with this y true value

1. What is the output should shape with?
   




2. Which shape do I have to reshape this y true value? and how?
   

---

Dataset is from Leeds Sports Dataset (LSP): 11000 training and 1000 testing images from sports activities with challenging in terms of appearance and especially articulations. The majority of people have 150 pixel height. For each person the full body is labeled with total 14 joints.

python deep-learning keras cnn

share|improve this question

asked Apr 6 at 12:47

kim Yumikim Yumi

62

$endgroup$

image shape = (105, 157, 3)

y=array([[145.82985678, 95.5022898 , 1. ],
 [111.87785389, 83.38437237, 1. ],
 [ 0. , -27.29260682, 0. ],
 [146.22287032, 67.62016267, 1. ],
 [114.716285 , 89.0394005 , 1. ],
 [147.03073148, 90.65512282, 1. ],
 [ 0. , -27.29260682, 0. ],
 [ 0. , -27.29260682, 0. ],
 [ 0. , -27.29260682, 0. ],
 [102.99138111, 63.60269094, 1. ],
 [107.44553454, 91.85599752, 1. ],
 [112.68571505, 118.12240233, 1. ],
 [ 95.72063065, 64.38871802, 1. ],
 [ 76.74681039, 60.37124629, 1. ]])

share|improve this question

asked Apr 6 at 12:47

kim Yumikim Yumi

62

share|improve this question

asked Apr 6 at 12:47

kim Yumikim Yumi

62

asked Apr 6 at 12:47

kim Yumikim Yumi

62

asked Apr 6 at 12:47

kim Yumikim Yumi

62