NEED HELP

5 replies
winston
Joined: 10/02/2009
User offline. Last seen 22 weeks 6 days ago.

Need help to fix this problem this problem, problem is the result after installation.

rizalzaf
rizalzaf's picture
Joined: 12/22/2008
User offline. Last seen 8 weeks 3 days ago.
Re: Need Help

Could you desscribe the detail of the problem please. What the problem is, when it occured, what the error message given by the application, etc. It will make us easier to analyze the problem, and what cause the problem.

Thank you.

__________________

Rizal Zaini Ahmad Fathony
Zaitun Time Series Developer Team

Paco (not verified)
Question

How can add observations in a saved project?
For example: I have a project with ten observations betwen 1998 and 2008 (annual).
Now, I want to add year 2009, but i don`t find like "Edit Project" or something

rizalzaf
rizalzaf's picture
Joined: 12/22/2008
User offline. Last seen 8 weeks 3 days ago.
Re: Question

Hi Paco,

Thank you for your question.

At this version, Zaitun Timt Series still doesn't support project editing feature.
We will add this to our to do list in developing Zaitun Time Series.

Thank you for your feedback.

__________________

Rizal Zaini Ahmad Fathony
Zaitun Time Series Developer Team

BL (not verified)
Neural Network

Hi,
why Zaitun calculating/forecasting different value for 2 or 3 or more independent calculation with
same input data.Results must be the same-identical?
Please react and answer,
Regards,
bl

rizalzaf
rizalzaf's picture
Joined: 12/22/2008
User offline. Last seen 8 weeks 3 days ago.
Re: Neural Network

Hi, BL

Thank you for your question.

The result of Neural Network modeling depends on these factors:
- Network topology: how much the layers and neurons, what is the activation function used, etc
- Learning process: Learning algorithm, stooping criteria, initializing weight, etc.

The process of initializing weight in learning process of a neural network is done by giving the random value to each weight in neural network topology. The random value can be different for each try. Therefore, the result of the neural network modeling can be different for each try. Another reason is the different stopping criteria lead to the different result of neural network modeling.

But if you use another method available in Zaitun Time Series, the result are same identical for each try.

Thank you. Smile

__________________

Rizal Zaini Ahmad Fathony
Zaitun Time Series Developer Team

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.
  • You may quote other posts using [quote] tags.
  • Textual smileys will be replaced with graphical ones.

More information about formatting options