WreckItTim's profile - activity

Ya I ended up just making my own mlp from scratch haha, I need the control. It’s just excel dude

sjhalayka I am really confused why you gave me your code. This is a problem dealing with OpenCV?

Hey LBerger I rewrote the question to be more clear what the problem is.

ANN_MLP using UPDATE_WEIGHTS to graph error - broken I ran into this problem while trying to make a learning curve of an

ANN_MLP using UPDATE_WEIGHTS to graph error - broken I ran into this problem while trying to make a learning curve of an

ANN_MLP using UPDATE_WEIGHTS to graph error - broke Doing some benchmarks and I want to make learning curves, showing er

ANN_MLP using UPDATE_WEIGHTS to graph error - broke Doing some benchmarks and I want to make learning curves, showing er

ANN_MLP using UPDATE_WEIGHTS to graph error vs number of training epochs Doing some benchmarks and I want to make learni

ANN_MLP UPDATE_WEIGHTS wonky Doing some benchmarks and I want to make learning curves, showing error versus number of tr

Yes it’s a regression problem. The training data is still the same for both network1 and network2. Shouldn’t the random

Yes it’s a regression problem. berak contionous responses are fine, I think one of us confused? Could be me. The traini

Yes it’s a regression problem. berak contionous responses are fine, I think one of us confused? Could be me. The traini

Shuffling data differently shouldn’t have that big of an effect on how quickly the error converges?

What would be the point of trying the XOR problem?

It’s the same results if I comment out the random seed lines.

ANN_MLP UPDATE_WEIGHTS wonky Doing some benchmarks and I want to make learning curves, showing error versus number of tr

ANN_MLP UPDATE_WEIGHTS wonky Doing some benchmarks and I want to make learning curves, showing error versus number of tr

Thanks, I posted the issue on GitHub!

Thanks, I posted the issue of GitHub!

termination criteria broken if you want to specify number of epochs cv::Ptr<cv::ml::ann_mlp> network = cv::ml::ANN

Ya thanks dude. I found the issue, the termination criteria is BROKEN on OpenCV, See this line: https://github.com/openc

I am not having any issues with number of nodes/layers. I am just using two layers with 40 nodes each. The problem is wh

Ya my job right now is to determine if this problem is solvable with an MLP, sorry I can't talk about it too much becaus

Hello dev team,

I had some general questions on max MLP parameters: Max number of layers? Max number of nodes per layer (and is this dependent on the number of layers used)? Max number of epochs you can set for training termination criteria?

I know these numbers are directly limited by the size of the variable type used, but curious in implementation what the actual limits are.

Thanks! -Tim

Thanks for the links to source code and response! From the train method here: https://github.com/opencv/opencv/blob/978

Max ANN_MLP size? Hello dev team, I had some general questions on max MLP parameters: Max number of layers? Max number

Max ANN_MLP size? Hello dev team, I had some general questions on max MLP parameters: Max number of layers? Max number

Okay so tanh(x) = sigmoid function = (1+e^(-2x)) / (1-e^(-2x)) so the activation functions are the same. The sigmoid fun

Okay so tanh(x) = sigmoid function = (1+e^(-2x)) / (1-e^(-2x)) so the activation functions are the same. The sigmoid fun

Okay so tanh(x) = sigmoid function = (1+e^(-2x)) / (1-e^(-2x)) so the activation functions are the same. The sigmoid fun

Ya you nailed it. Smart thinking diving into the source code. I posted an answer to this.

In the docs, http://docs.opencv.org/2.4/modules/ml/doc/neural_networks.html The sigmoid function is described in the top

MLP Activation Function - train vs predict Hey, so my question was for the developers. In the docs, it says that train