Added Transfer learning (#29462)
* Added Transfer learning * Added VGGnet * Rename guide/english/machine-learning/deep-learning/transfer-learning/VGGnet.md to guide/english/machine-learning/deep-learning/transfer-learning/vggnet/index.md
This commit is contained in:
@ -0,0 +1,26 @@
|
|||||||
|
---
|
||||||
|
title: Transfer learning
|
||||||
|
---
|
||||||
|
|
||||||
|
### What is Transfer Learning ?
|
||||||
|
|
||||||
|
Transfer learning is the re-use of the pre-trained models on a different problem. It is very popular in Deep Learning as features learnt by one model trained on a Huge dataset can be re-used for another problem which as a small dataset to train on.
|
||||||
|
|
||||||
|
### Pretrained models
|
||||||
|
|
||||||
|
Some of the pretrained models that are widely used are :
|
||||||
|
* VGG16
|
||||||
|
* Xception
|
||||||
|
* VGG19
|
||||||
|
* ResNet50
|
||||||
|
* InceptionV3
|
||||||
|
* InceptionResNetV2
|
||||||
|
* MobileNet
|
||||||
|
* DenseNet
|
||||||
|
* NASNet
|
||||||
|
* MobileNetV2
|
||||||
|
|
||||||
|
### More Information
|
||||||
|
* Keras Pretrained Models https://keras.io/applications/
|
||||||
|
* Pre Trained Model depot https://modeldepot.io/
|
||||||
|
* Datacamp article https://www.datacamp.com/community/tutorials/transfer-learning
|
@ -0,0 +1,12 @@
|
|||||||
|
---
|
||||||
|
title: VGGnet
|
||||||
|
---
|
||||||
|
VGGnet are CNN models created by research group Visual Graphic Group as a winning entry to the ImageNet Large Scale Visual Recognition Competition (ILSVRC) in 2014 in the Localization category and the runner up in Image Classification category in the same competion.
|
||||||
|
|
||||||
|
VGG group have released the following models:
|
||||||
|
* 16 Layer model - [Information](https://gist.github.com/ksimonyan/211839e770f7b538e2d8#file-readme-md)
|
||||||
|
* 19 Layer model - [Information](https://gist.github.com/ksimonyan/3785162f95cd2d5fee77#file-readme-md)
|
||||||
|
|
||||||
|
### More Information
|
||||||
|
* Visual Geometry Group website http://www.robots.ox.ac.uk/~vgg/research/very_deep/
|
||||||
|
|
Reference in New Issue
Block a user