Skip to content

Latest commit

 

History

History
74 lines (66 loc) · 2.36 KB

README.md

File metadata and controls

74 lines (66 loc) · 2.36 KB

AN2DL-Challenge

These challenges have been proposed by the course Artificial Neural Networks and Deep Learning of polimi.

In these challenges/projects we have been asked to perform image classification (challenge 1) and data forecasting (challenge 2) using Neural Networks and Deep Learning.

Remarks about challenge 1

After two weeks of development we have been able to achieve an accuracy, on a hidden dataset, of 94.15%.

The task

In this homework you are required to classify images of leaves, which are divided into categories according to the species of the plant to which they belong. Being a classification problem, given an image, the goal is to predict the correct class label.

The data

  • Image size: 256x256
  • Color space: RGB (read as 'rgb' in ImageDataGenerator.flow_from_directory ('color_mode' attribute) or use PIL.Image.open('imgname.jpg').convert('RGB'))
  • File Format: JPG
  • Number of classes: 14
  • Classes: 0. "Apple"
    1. "Blueberry"
    2. "Cherry"
    3. "Corn"
    4. "Grape"
    5. "Orange"
    6. "Peach"
    7. "Pepper"
    8. "Potato"
    9. "Raspberry"
    10. "Soybean"
    11. "Squash"
    12. "Strawberry"
    13. "Tomato"
  • images per class:
    • Apple : 988
    • Blueberry : 467
    • Cherry : 583
    • Corn : 1206
    • Grape : 1458
    • Orange : 1748
    • Peach : 977
    • Pepper : 765
    • Potato : 716
    • Raspberry : 264
    • Soybean : 1616
    • Squash : 574
    • Strawberry : 673
    • Tomato : 5693

Remarks about challenge 2

We have been able to achieve a RMSE (Root Mean Squared Error) of 3.7108 overall on a hidden test set.

The task

In this homework, you are required to predict future samples of a multivariate time series. The goal is to design and implement forecasting models to learn how to exploit past observations in the input sequence to correctly predict the future.

The data

  • Multivariate time series with the following characteristics:
    • Length of the time series (number of samples in the training set): 68528
    • Number of features: 7
    • Name of the features:
      • 'Sponginess'
      • 'Wonder level'
      • 'Crunchiness'
      • 'Loudness on impact'
      • 'Meme creativity'
      • 'Soap slipperiness'
      • 'Hype root'
    • Uniform sampling rate.

Authors