change activation function but return same accuracy value!!!

[bakigkgz1]

Hello, when I change the activation function inside the formula.cpp file (for example, when I write the formula for the tanh function), it consistently returns an accuracy value of 9.80%. Do you know why this might be happening?

Thank you for help

[leo-c-ling99]

Hi,

Disclaimer it's been a few year since I've looked at this code. It seems that the gradient for the activation is hardcoded within train.cpp. You might need to update their backpropagation code if you want to change the activation.

MLP_NeuroSim_V3.0/Train.cpp

Line 507 in cc13863

// Backpropagation

[bakigkgz1]

Thank you Leo Ling ***@***.***>, 19 Ara 2023 Sal, 20:24 tarihinde şunu yazdı:

…

Hi, Disclaimer it's been a few year since I've looked at this code. It seems that the gradient for the activation is hardcoded within train.cpp. You might need to update their backpropagation code if you want to change the activation. https://github.com/neurosim/MLP_NeuroSim_V3.0/blob/cc1386372d01fc022a9bf52cabd8c96e94fb838b/Train.cpp#L507 — Reply to this email directly, view it on GitHub <#11 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AI6Q7AWOPYXIRHU4JZRCSQ3YKHEVNAVCNFSM6AAAAABAYNB7HWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQNRTGE4TKMZUGE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

-- Ögr.Gör.Baki GÖKGÖZ Gümüşhane Üniversitesi www.bakigokgoz.com ***@***.***

[bakigkgz1]

Hello, first of all, thank you for your reply. I changed the part you said as follows, but after a certain iteration, it gets stuck at 10.35%. (I want to use tanh as activation function)

// Backpropagation
/* Second layer (hidden layer to the output layer) */

		for (int j = 0; j < param->nOutput; j++){
                             s2[j] = -2*a2[j] * (1 - a2[j]*a2[j])*(Output[i][j] - a2[j]);
		}

		/* First layer (input layer to the hidden layer) */
		std::fill_n(s1, param->nHide, 0);
		#pragma omp parallel for
					
		for (int j = 0; j < param->nHide; j++) {
			for (int k = 0; k < param->nOutput; k++) {
				s1[j] += a1[j] * (1 - a1[j]*a1[j]) * weight2[k][j] * s2[k];
			}
		}

[bakigkgz1]

Hello again, I apologize for the disturbance. In this application, we want to use different activation functions and for this purpose, we are looking to hire someone to do this job for a certain fee. Can you or someone else help with this? Thank you and best regards. Baki gökgöz ***@***.***>, 19 Ara 2023 Sal, 23:09 tarihinde şunu yazdı:

…

Thank you Leo Ling ***@***.***>, 19 Ara 2023 Sal, 20:24 tarihinde şunu yazdı: > Hi, > > Disclaimer it's been a few year since I've looked at this code. It seems > that the gradient for the activation is hardcoded within train.cpp. You > might need to update their backpropagation code if you want to change the > activation. > > > https://github.com/neurosim/MLP_NeuroSim_V3.0/blob/cc1386372d01fc022a9bf52cabd8c96e94fb838b/Train.cpp#L507 > > — > Reply to this email directly, view it on GitHub > <#11 (comment)>, > or unsubscribe > <https://github.com/notifications/unsubscribe-auth/AI6Q7AWOPYXIRHU4JZRCSQ3YKHEVNAVCNFSM6AAAAABAYNB7HWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQNRTGE4TKMZUGE> > . > You are receiving this because you authored the thread.Message ID: > ***@***.***> > -- Ögr.Gör.Baki GÖKGÖZ Gümüşhane Üniversitesi www.bakigokgoz.com ***@***.***

-- Ögr.Gör.Baki GÖKGÖZ Gümüşhane Üniversitesi www.bakigokgoz.com ***@***.***