https://github.com/jeugregg/deeplearningpicturedogs
Classify dogs pictures by Deep Learning CNN neural networks
https://github.com/jeugregg/deeplearningpicturedogs
classez-des-images cnn-keras data data-science ipynb neural-network vision
Last synced: 11 months ago
JSON representation
Classify dogs pictures by Deep Learning CNN neural networks
- Host: GitHub
- URL: https://github.com/jeugregg/deeplearningpicturedogs
- Owner: jeugregg
- Created: 2019-11-10T16:01:43.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2020-03-08T22:56:09.000Z (over 6 years ago)
- Last Synced: 2025-03-02T06:14:38.669Z (over 1 year ago)
- Topics: classez-des-images, cnn-keras, data, data-science, ipynb, neural-network, vision
- Language: Jupyter Notebook
- Homepage:
- Size: 20.5 MB
- Stars: 0
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# DeepLearningPictureDogs
Classify pictures with Deep Learning Model
This study is to learn to use, create and train deep learning CNN models (ResNet / VGG).
An association which helps and hosts a lot of dogs need to classify pictures of their animals.
To do that, it is possible to use CNN models by training on Stanford Dogs Dataset (http://vision.stanford.edu/aditya86/ImageNetDogs/).
These is my works about this study :
01_OC_IML_Project_6_explorations.ipynb : exploration
02_OC_IML_Project_6_my_neural_network.ipynb & 03_OC_IML_Project_6_my_neural_network_2.ipynb : Creation / test of a neural network CNN from scratch.
04_OC_IML_Project_6_10breeds.ipynb & 05_OC_IML_Project_6_3breeds.ipynb : VGG-16 Neural Network transfert learning
06_OC_IML_Project_6_3breeds_ResNet50.ipynb & 07_OC_IML_Project_6_10breeds_ResNet50.ipynb : ResNet-50 transfert learning
08_OC_IML_Project_6_my_neural_network_DAugm_01.ipynb : data augmentation on my neural network from scratch
09_OC_IML_Project_6_10breeds_ResNet50_script.ipynb : notebook to create and test script
- script_predict_breed.py : command line script to predict breed (ResNet50 10 breeds)
- ResNet50_TL_10b_9.h5 : transfert learning model
- df_dogs.pkl : contains list of breeds (and other stuff about training pictures)