https://github.com/muizzer07/model-trainer
A Django Webapp to build AI Image Classifier Models (Train, infer, download weight files) without coding
https://github.com/muizzer07/model-trainer
deep-learning django image-classifier pytorch sklearn
Last synced: about 2 months ago
JSON representation
A Django Webapp to build AI Image Classifier Models (Train, infer, download weight files) without coding
- Host: GitHub
- URL: https://github.com/muizzer07/model-trainer
- Owner: MuizZer07
- Created: 2019-07-29T00:30:34.000Z (almost 6 years ago)
- Default Branch: master
- Last Pushed: 2019-08-05T12:42:49.000Z (almost 6 years ago)
- Last Synced: 2025-01-28T04:25:26.664Z (4 months ago)
- Topics: deep-learning, django, image-classifier, pytorch, sklearn
- Language: JavaScript
- Homepage:
- Size: 586 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
### A Django Webapp to build AI Image Classifier Models without coding
#### Implementations:
- Upload data (only csv formats contains images, labels collumn, images must have 3 dimesions)
- Data Loading and Preprocessing
- Transfer Learning (resnet, vgg, alexnet, inception, densenet, squeezenet
- Save models
- User authentication, private, public data
#### TO DO:
- Showing live logs during training
- Store training History, create user dashboard
- Create inference for each model
- Data format independent platform
- Data preprocessing, training user options (hyper-parameters tunining options)
- Implement more machine learning algorithms, generalized cnn architectures
##### ** Limitations:
- Limited input formats, Can only take csv files, images with 3 dimensions
- Training resource, Storage resource, Time complexity etc.