Dear Mr. Thomas, I am very novice on this software. Looking at this tutorial on optimization parameter, I have a question regarding the optimization of hydrology model. Currently, I am working on Snowmelt Runoff Model. For executing this model, the study area should be divided into zone( elevation zone), and for every zone, we have input data : Daily temperature reading, daily precipitation and daily snow cover. Along with this input data, I have around 7 parameters which has a certain range and these parameters are use to calibrate the model. However, it could be done by trail and error method to get the optimum parameters for every zone ( daily parameters), but the method is very tedious and time consuming. I would like ask whether it is possible to find out daily optimum parameters of the above mentioned model. The model performance is based on Nash Sutcliffe coefficient ( Comparing computed runoff against measure daily runoff/discharge ). Hope I made myself clear with my doubt. Thanking you.
This process worked form me, when using Decision Tree rather than Neural Network. BUT, I cant see the details of the actual Decision Tree model, I can only see its performance measure! How can I access the description of the best performing DT model, as it has been optimised?
Hello Mr. Thomas, I want to ask, how to optimize algorithms Decision tree algorithm using Particle Swarm Optimization in RapidMiner? thank you for your willingness to share with my host.
The log table doesnt appear for me while the table is building. but I have 4000++ example and around 1000 attributes. My computers memory is 4GB Is that the reason?
in my videos /watch?v=L5lwpypsyuQ and /watch?v=BCn35mWsZ0U I feature future-prices of gold and silver, a range of values for daily closing prices weeks in advance. I would like to automate the process for any kind of stock, do you know how to do this in rapid-miner? I need it to auto-load data-sets and to find correlations & apply models to generate future prices all overnight or while I'm busy doing stuff. Can it?
Exactly what I was looking for, thanks for detailed tutorial!
You're very welcome.
Thanks a lot for your tutorial !!! This one helped me a lot !
Hello sir, can you explain a tutorial regarding the applications of the algorithm Particle Swarm Optimization in RapidMiner, please.
Dear Mr. Thomas,
I am very novice on this software. Looking at this tutorial on optimization parameter, I have a question regarding the optimization of hydrology model.
Currently, I am working on Snowmelt Runoff Model. For executing this model, the study area should be divided into zone( elevation zone), and for every zone, we have input data : Daily temperature reading, daily precipitation and daily snow cover.
Along with this input data, I have around 7 parameters which has a certain range and these parameters are use to calibrate the model. However, it could be done by trail and error method to get the optimum parameters for every zone ( daily parameters), but the method is very tedious and time consuming.
I would like ask whether it is possible to find out daily optimum parameters of the above mentioned model. The model performance is based on Nash Sutcliffe coefficient ( Comparing computed runoff against measure daily runoff/discharge ).
Hope I made myself clear with my doubt.
Thanking you.
Hello! Thanks for the video. Could you tell my why "hidden_layers" parameters is gray out in the "Select Parameters" screen?
This process worked form me, when using Decision Tree rather than Neural Network. BUT, I cant see the details of the actual Decision Tree model, I can only see its performance measure! How can I access the description of the best performing DT model, as it has been optimised?
can you choose nominal instead of numerical x validation?
This method would check the performance, but how to check the accuracy?
@ytgv3fc7 Yes it can do that but you need to set up the Rapid Analytics server and then set a CRON job to run the processes.
Hello Mr. Thomas,
I want to ask,
how to optimize algorithms Decision tree algorithm using Particle Swarm Optimization in RapidMiner?
thank you for your willingness to share with my host.
The log table doesnt appear for me while the table is building. but I have 4000++ example and around 1000 attributes. My computers memory is 4GB Is that the reason?
in my videos /watch?v=L5lwpypsyuQ and /watch?v=BCn35mWsZ0U I feature future-prices of gold and silver, a range of values for daily closing prices weeks in advance. I would like to automate the process for any kind of stock, do you know how to do this in rapid-miner? I need it to auto-load data-sets and to find correlations & apply models to generate future prices all overnight or while I'm busy doing stuff. Can it?