Skip to content

Latest commit

 

History

History
10 lines (6 loc) · 634 Bytes

File metadata and controls

10 lines (6 loc) · 634 Bytes

This is edited notebook from the course Practical deeplearning for coders.

The main goal of this notebook is to take advantage of capabilities of existing pretrained models. To conclude in a single word - 'Transfer Learning'

The model used is pretrained 'Resnet18', finetuned on images of target classes. The images are scrapped from internet.

The model took around 3 minutes to train on kaggle 'CPU'.

The inference app is hosted on Huggingface Space.