Learn Before
Number of Layers to Freeze When Using a Pre-trained Model for Transfer Deep Learning
The decision of how many layers to freeze or replace when using a pre-trained model for transfer learning depends on the size of the dataset:
- Small dataset: Replace only the last layer.
- Large dataset: Replace multiple of the last layers with new layers.
- Extremely large dataset: If sufficient computational power is available, the pre-trained network can be used for weight initialization, and all the weights across the entire network can be trained.
0
1
Contributors are:
Who are from:
Tags
Data Science
Related
POPULAR PRE-TRAINED MODELS
Number of Layers to Freeze When Using a Pre-trained Model for Transfer Deep Learning
Implementing Freezing Layers When Using a Pre-trained Model for Transfer Deep Learning
An engineer needs to build a model to classify 15 types of local wildflowers using a custom dataset of only 900 images. They select a very deep and complex neural network that was previously trained on a dataset of over a million general-purpose images (e.g., animals, vehicles, household objects). The engineer's strategy is to retrain all layers of this complex network from scratch, using only their small wildflower dataset. What is the most likely outcome of this strategy?
You are tasked with building an image classifier for a new, specialized task (e.g., identifying specific types of industrial equipment), but you only have a small, custom dataset. You decide to adapt a model that has already been trained on a very large, general image dataset. Arrange the following steps in the correct logical order to implement this strategy.
Adapting a Pre-trained Network for a New Task
Hot Dog Recognition Example in Fine-Tuning
Computational Benefits of Freezing Pretrained Parameters