Learning Machine Algorithms
- or -
Post a project like this$293
- Posted:
- Proposals: 7
- Remote
- #2109609
- Expired
Professional Graphic designer, photo editor, clipping path, images background removes, Photo cleanup/ touch up, Image Manipulation
Medina
352228192882619624052259406227525122843252346404
Description
Experience Level: Expert
I´m looking for a secon opinion and solutions about my learning machine algorithms.
It´s about predictions on roulette casinos.
Do you know the roulette game?
I have a developed and trained machine learning (Tensorflow and Keras) to predict the results of the roulette, is based on a recurrent NN (LSTM) with certain postulates and data that I contributed; but it has the defect that it seems to understand that the number "1" is safe and appears with an abnormal frequency in said predictions.
The NN feeds on objects (Black / Red, High / Low, etc ...), until arriving at the prediction of the stright number, and not vice versa, that is to directly predict the next figure.
The code shows me each of all the figures most likely to appear, including the direct number.
As you can see, the results are all the figures of roulette, including numbers.
All work is in Python exclusively, and is aimed at making a later .exe with an assembler that collaborates with me.
I would like to get not only a second opinion, but a solution or correction that will improve the current work that I already have.
Then, It will be needed and contributed a test about rate of hits over the figure of straight numbers, and It will be over a series of 100 and other of 300 numbers.
It´s about predictions on roulette casinos.
Do you know the roulette game?
I have a developed and trained machine learning (Tensorflow and Keras) to predict the results of the roulette, is based on a recurrent NN (LSTM) with certain postulates and data that I contributed; but it has the defect that it seems to understand that the number "1" is safe and appears with an abnormal frequency in said predictions.
The NN feeds on objects (Black / Red, High / Low, etc ...), until arriving at the prediction of the stright number, and not vice versa, that is to directly predict the next figure.
The code shows me each of all the figures most likely to appear, including the direct number.
As you can see, the results are all the figures of roulette, including numbers.
All work is in Python exclusively, and is aimed at making a later .exe with an assembler that collaborates with me.
I would like to get not only a second opinion, but a solution or correction that will improve the current work that I already have.
Then, It will be needed and contributed a test about rate of hits over the figure of straight numbers, and It will be over a series of 100 and other of 300 numbers.
Esferasoft Solutions Pvt Ltd
100% (1)Projects Completed
1
Freelancers worked with
1
Projects awarded
0%
Last project
29 Jul 2018
India
New Proposal
Login to your account and send a proposal now to get this project.
Log inClarification Board Ask a Question
-
There are no clarification messages.
We collect cookies to enable the proper functioning and security of our website, and to enhance your experience. By clicking on 'Accept All Cookies', you consent to the use of these cookies. You can change your 'Cookies Settings' at any time. For more information, please read ourCookie Policy
Cookie Settings
Accept All Cookies