What must i understand with run results? #718
-
|
I trained a model with a dataset by net = new brain.NeuralNetwork(); and then run with prediction data. X: 1.9106081161623933e-9 Prediction results as above. As you can see, both Y and Z are nearly 1. What must i understand with this results? |
Beta Was this translation helpful? Give feedback.
Replies: 3 comments 3 replies
-
|
What is the training data? |
Beta Was this translation helpful? Give feedback.
-
|
Training data has alot of items. A sample as below: input: output: |
Beta Was this translation helpful? Give feedback.
-
|
If you want me to share full data, i can share. |
Beta Was this translation helpful? Give feedback.
-
May i wait for reply or not? |
Beta Was this translation helpful? Give feedback.
-
|
Hi Mr. Robert. Is this results normal or not? |
Beta Was this translation helpful? Give feedback.
-
|
You'll want to normalize your data on the same scale. Likely between 0 and 1 for each value. You can later de-normalize it (or scale it). The problem you're going to run into is that a input that is very tiny is like a whisper into your ear, and a large number is like yelling. By putting all your data in a scale between 0 and 1 it allows the network to better understand what's going on. This is a principle throughout machine learning. |
Beta Was this translation helpful? Give feedback.
You'll want to normalize your data on the same scale. Likely between 0 and 1 for each value. You can later de-normalize it (or scale it). The problem you're going to run into is that a input that is very tiny is like a whisper into your ear, and a large number is like yelling. By putting all your data in a scale between 0 and 1 it allows the network to better understand what's going on. This is a principle throughout machine learning.