Jovian
Sign In
Learn practical skills, build real-world projects, and advance your career
!pip install jovian --upgrade --quiet
import jovian
# Execute this to save new versions of the notebook
jovian.commit(project="final-dl-project-example")
[jovian] Detected Colab notebook... [jovian] Uploading colab notebook to Jovian...
[jovian] Error: Looks like the notebook is missing output cells, please save the notebook and try jovian.commit again.
[jovian] Capturing environment.. [jovian] Committed successfully! https://jovian.ai/alparslantamermain/final-dl-project-example 
'https://jovian.ai/alparslantamermain/final-dl-project-example'

Picking a good dataset

  • Should be toy/standard dataset (MNIST, Titanic, CIFAR)
  • Should be large enough (> 1000 images)
  • Sholud have enough variety
  • It should be possible to build a good model
  • Should not be too large (< 5GB)
    • Exception: pick a 10% or 1% sample

Dataset Candidates

CelebA Dataset: https://www.kaggle.com/jessicali9530/celeba-dataset

Flower Recognition Dataset: https://www.kaggle.com/alxmamaev/flowers-recognition

Identifying Flower Species using Deep Learning ad Pytorch

TODO - Introduction

We are going to do it in the following steps:

  1. Pick a dataset
  2. Download the dataset
  3. Import the dataset using Pytorch
  4. Explore the dataset
  5. Prepare the dataset for training
  6. Move the dataset to the GPU
  7. Define NN
  8. Train the model
  9. Make predictions on sample images

Iterate on it with different networks & hyperparameters.