All Questions
51 questions
0
votes
0
answers
131
views
Batch normalization in neural network for regression task
I have this neural network that I want to fit for regression. When I don't use batch normalization between layers it performs good, but when I add batch normalization it performs really bad.
For ...
- 117
1
vote
0
answers
32
views
Multioutput regression with differing results when explicitly creating individual output neurons
I have a neural network model defined as such:
class FDC_Model(Model):
def __init__(self, out_names):
super(FDC_Model, self).__init__()
self.out_names = out_names
self....
- 172
0
votes
0
answers
107
views
How to find the predicted values with Keras
I'm learning keras, and would like to see the predicted numbers that are returned. The model has a number of items returned, but none of them seem to be the predicted values.
df <- MASS::Boston
...
- 360
2
votes
1
answer
205
views
LSTM Regression issues with masking and intuition (keras)
I am using this architecture (a Masking Layer for varying trajectory lengths that are padded with 0s to maximum length trajectory followed by a LSTM with a dense layer afterwards that outputs 2 values)...
- 284
0
votes
0
answers
408
views
Prediction results in negative values
I am using a simple neural network to perform multiple regression but the prediction results in negative values for all prediction values that should be 0.
What should I do? Is it okay to change the ...
- 31
0
votes
0
answers
50
views
Why is the score of the tuned model of mine obviously non-optimal (negative)?
I am trying to tune a stacking regressor which includes a layer constituting a decision tree, a random forest, and a deep network, and a xgbregressor as a blender.
My tuning procedure below
import os
...
user1393214
0
votes
0
answers
775
views
MinMaxScaler for predictions after training
Could someone give me a tip on how to use Scikit MinMaxScaler when predicting with an MLP neural network?
I know this part of the code at the very end isn't right, the data used to #make single ...
- 3,530
0
votes
1
answer
279
views
CNN regression results in two distinct (incorrect) predictions
I'm trying to solve a regression problem using a Python Keras CNN (Tensorflow as the backbone), where I try to predict a single y-value based on an 8-dimensional satellite image (23x45 pixels) that I ...
3
votes
2
answers
1k
views
Keras Sequential Model Non-linear Regression Model Bad Prediction
To test a nonlinear sequential model using Keras,
I made some random data x1,x2,x3
and y = a + b*x1 + c*x2^2 + d*x3^3 + e (a,b,c,d,e are constants).
Loss is getting low really quickly but the model ...
- 63
-1
votes
1
answer
29
views
NN Keras Regression: Different Input for Training and Prediction [closed]
I want to achieve a sports prediction where the (test) input for a model is only 1 (ID) out of 20 variables (that training gets)
I build a NN Regression model in python with keras that uses ~ 20 input ...
- 166
0
votes
0
answers
133
views
Regression with Keras or what model should i use?
I am lost and have no idea if CNN is what I need.
I have a small .csv dataset of soccer statistics (~3,5k rows and 28 columns).
p.E.:
players; min; goal; foul; ... tackle
14; 1004; 0; 12; ...
- 166
6
votes
5
answers
8k
views
Machine learning regression model predicts same value for every image
I am currently working on a project involving training a regression model, saving it and then loading it to make further predictions using that model. However I'm having a problem. Each time that I ...
- 497
1
vote
0
answers
368
views
Non linear regression using neural networks: Curve not fitting at some points
My model has 2 independent variables and one dependent variable.
I am using a neural network 5 layer neural network to fit my curve, and I am using Keras for the purpose.
After 2000 epochs, I am ...
- 53
0
votes
1
answer
2k
views
Regression in Keras has low MSE but results are way off
[Update: An answer by desertnaut below caught a coding mistake (doubly un-scaling my predictions, leading them to be ~ E+22), but even after correcting this, my problem remains.]
I'm struggling to ...
- 26.1k
1
vote
1
answer
372
views
Unable to get good results when trying to predict mean as well as standard deviation in a image regression task
I am trying to regress two variables (mean and std) and then trying to optimize log(gaussian_distribution) = log(std) + (target - mean) / (2 * std ^ 2).
Note that on the same data, if change loss to (...
