'Deep learning result graph is is limited to the average area
]
You can check the Network Model and Result from the Photos.
Result datas are stuck in the "average band" and can't forecasting the exact value.
I used a 3years data to training(train, val = 8:2) and 1 year of dataset is for Test.
change Scaler, model, dense, activation function, batch size, optimizer but nothing couldn't solved graph's problem.
Does anyone know the solutions?
Appreciate any suggestions. Thx for reading!
Solution 1:[1]
I think it is not about the model that is some DATA need to be re-format that will help models understanding better. Forgetting about the model, any model can learn in their scopes I also using only 32 cells of LSTM but the DATA need tobe easy understanding. It is when the DATA is photo, see the minimum and maximum values you don't need to adjusting to average values but target result you expected ! Here below we using just only greater to filtered out background and remains the player and emermies !
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|---|
| Solution 1 | Martijn Pieters |