In Deep learning how much intermidate layers consists how to decide
How much deep learning intermidate layers consists for getting desirable outcome, or it's data of eg 10000records of excel file.
1 answers ( 0 marked as helpful)
Super learner
This user is a Super Learner. To become a Super Learner, you need to reach Level 8.
It depends on:
Problem complexity
Dataset size
Feature richness
For 10,000 data records (Excel):
You do NOT need a deep network.
Typically 2–3 hidden layers with 32–128 neurons each is enough.
Too many layers = overfitting (since 10k rows is a small dataset).
General Guideline:
Start with a simple model.
Use regularization (dropout, early stopping) if adding more layers.
For many small datasets, traditional ML (XGBoost, Random Forest) can perform better than deep learning.
Problem complexity
Dataset size
Feature richness
For 10,000 data records (Excel):
You do NOT need a deep network.
Typically 2–3 hidden layers with 32–128 neurons each is enough.
Too many layers = overfitting (since 10k rows is a small dataset).
General Guideline:
Start with a simple model.
Use regularization (dropout, early stopping) if adding more layers.
For many small datasets, traditional ML (XGBoost, Random Forest) can perform better than deep learning.
