An open API service indexing awesome lists of open source software.

https://github.com/shasan7/dl_fine_tuning_matlab

Fine-Tune Pretrained Image Classification Models in MATLAB
https://github.com/shasan7/dl_fine_tuning_matlab

alexnet computer-vision darknet deep-learning densenet201 efficientnetb0 fine-tuning fine-tunning googlenet image-classfication image-classification inceptionresnetv2 inceptionv3 matlab mobilenetv2 resnet squeezenet transfer-learning vgg xception

Last synced: 8 months ago
JSON representation

Fine-Tune Pretrained Image Classification Models in MATLAB

Awesome Lists containing this project

README

          

# Fine-Tune Pretrained Deep Learning Image Classification Models in MATLAB

While modifying these scripts for use, the dataset location is supposed to contain **subfolders with folder names as labels of corresponding image categories**.
Due to different final layers structure for different networks, i had to provide **separate codes** for each of them.
In my MATLAB version (R2023a), I can only save the model with "best-validation" where the "Metric' is automatically assumed validation loss. If you version is compatible, try **"best-validation" with "accuracy" Metric in the trainingOptions for higher performance**.
If you require the **cross-validation** version of these codes, you'll find them in the **"Cross Validation"** folder. The corresponding **MATLAB installation packages** for pretrained networks are saved in **"MATLAB_DL_Packages"** folder.