We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
I inspected whole of the code, but I guess there is a mistake in this line. As far as I am concerned, the cost function used in your code is:
Therefore, If you are going to follow delta rule, the data needed for back-propagation is:
But in your code it is:
The difference between them is sigmoid function. I think there is a small mistake in this line.
recurrent_neural_net_demo/rnn.py
Line 77 in 35fb080
The text was updated successfully, but these errors were encountered:
No branches or pull requests
I inspected whole of the code, but I guess there is a mistake in this line.
![](https://render.githubusercontent.com/render/math?math=C%20%3D%20%20%5Cfrac%7B1%7D%7B2%7D%20(y%20-%20layer%5C_2)%5E2)
As far as I am concerned, the cost function used in your code is:
Therefore, If you are going to follow delta rule, the data needed for back-propagation is:
![](https://render.githubusercontent.com/render/math?math=%5CDelta%20W%20%3D%20%5Calpha%20(y%20-%20layer%5C_2)sigmoid%5C_output%5C_to%5C_derivative(np.dot(layer%5C_1%2Csynapse%5C_1)))
But in your code it is:
![](https://render.githubusercontent.com/render/math?math=%5CDelta%20W%20%3D%20%5Calpha%20(y%20-%20layer%5C_2)sigmoid%5C_output%5C_to%5C_derivative(layer%5C_2)))
The difference between them is sigmoid function.
I think there is a small mistake in this line.
recurrent_neural_net_demo/rnn.py
Line 77 in 35fb080
The text was updated successfully, but these errors were encountered: