All Projects → anubhavmaity → Sports Type Classifier

anubhavmaity / Sports Type Classifier

Classify the type of sports from images

Projects that are alternatives of or similar to Sports Type Classifier

Ucsandiegox Dse200x Python For Data Science
UCSandDiego Micro Masters Program
Stars: ✭ 27 (-3.57%)
Mutual labels:  jupyter-notebook
Textclassifier
tensorflow implementation
Stars: ✭ 944 (+3271.43%)
Mutual labels:  jupyter-notebook
Idb Idb Invest Coronavirus Impact Dashboard
Follow the impact of COVID-19 outbreak in Latin America in real time
Stars: ✭ 28 (+0%)
Mutual labels:  jupyter-notebook
Oxford Deepnlp 2017
🚀 🎉 ✨ Oxford Deep NLP 2017 Course Materials and Practicals, Solutions
Stars: ✭ 27 (-3.57%)
Mutual labels:  jupyter-notebook
Sklearn ensae course
Materials for a course on scikit-learn at ENSAE
Stars: ✭ 27 (-3.57%)
Mutual labels:  jupyter-notebook
Alfabattle2 1stproblem
Alfabattle 2.0 1st task Top-6 solution: 8-folds lgbm blend
Stars: ✭ 27 (-3.57%)
Mutual labels:  jupyter-notebook
Machine Learning Data Science Reuse
Gathers machine learning and data science techniques for problem solving.
Stars: ✭ 27 (-3.57%)
Mutual labels:  jupyter-notebook
Chexpert
CheXpert competition models -- attention augmented convolutions on DenseNet, ResNet; EfficientNet
Stars: ✭ 28 (+0%)
Mutual labels:  jupyter-notebook
Anatomyofmatplotlib
Anatomy of Matplotlib -- tutorial developed for the SciPy conference
Stars: ✭ 943 (+3267.86%)
Mutual labels:  jupyter-notebook
Advanced Gradient Obfuscating
Take further steps in the arms race of adversarial examples with only preprocessing.
Stars: ✭ 28 (+0%)
Mutual labels:  jupyter-notebook
Odsc east 2016
Stars: ✭ 27 (-3.57%)
Mutual labels:  jupyter-notebook
Sgdoptim.jl
A julia package for Gradient Descent and Stochastic Gradient Descent
Stars: ✭ 27 (-3.57%)
Mutual labels:  jupyter-notebook
Sid
Official implementation for ICCV19 "Shadow Removal via Shadow Image Decomposition"
Stars: ✭ 28 (+0%)
Mutual labels:  jupyter-notebook
Conjugate Gradient
Painless conjugate gradient notebooks
Stars: ✭ 27 (-3.57%)
Mutual labels:  jupyter-notebook
Data driven science python demos
IPython notebooks with demo code intended as a companion to the book "Data-Driven Science and Engineering: Machine Learning, Dynamical Systems, and Control" by J. Nathan Kutz and Steven L. Brunton
Stars: ✭ 27 (-3.57%)
Mutual labels:  jupyter-notebook
Mambo
Stars: ✭ 27 (-3.57%)
Mutual labels:  jupyter-notebook
Kispython
Курс программирования на языке Python
Stars: ✭ 27 (-3.57%)
Mutual labels:  jupyter-notebook
Medium Article
Repo for articles in my personal blog and Medium
Stars: ✭ 28 (+0%)
Mutual labels:  jupyter-notebook
Mask Rcnn Tensorflow
Fork of Tensorpack to make breaking performance improvements to the Mask RCNN example. Training is approximately 2x faster than the original implementation on AWS.
Stars: ✭ 28 (+0%)
Mutual labels:  jupyter-notebook
Pacmap
PaCMAP: Large-scale Dimension Reduction Technique Preserving Both Global and Local Structure
Stars: ✭ 28 (+0%)
Mutual labels:  jupyter-notebook

22 Types of Sports Classification

Following are the types of sports over which we are trying to classifying:

  • Swimming
  • Badminton
  • Wrestling
  • Olympic Shooting
  • Cricket
  • Football
  • Tennis
  • Hockey
  • Ice Hockey
  • Kabaddi
  • WWE
  • Gymnasium
  • Weight lifting
  • Volleyball
  • Table tennis
  • Baseball
  • Formula 1
  • Moto GP
  • Chess
  • Boxing
  • Fencing
  • Basketball

Sports Type

Data Distribution

Data has been downloaded with the help of gi2ds

Training set: 11524

Validation set: 2881

Model

Resnet-50

Data Augmentations

The following data augmentation has been applied to increase the no of images in the training set

  1. Flip horizontal
  2. Lighting
  3. Zooming
  4. Warping

Data Augmentations

Confusion Matrix

Confusion matrix

Pair of confused categories with minimum value of 2

[('motogp', 'formula1', 5),

('badminton', 'tennis', 4),

('weight_lifting', 'wrestling', 3),

('wrestling', 'wwe', 3),

('badminton', 'table_tennis', 2),

('basketball', 'volleyball', 2),

('boxing', 'wrestling', 2),

('hockey', 'ice_hockey', 2),

('kabaddi', 'hockey', 2),

('tennis', 'table_tennis', 2),

('weight_lifting', 'wwe', 2),

('wrestling', 'kabaddi', 2)]

Sample Images from the confused categories:

For some images you can see the prediction is correct but the manual labelling was done wrong. This shows the CNN are less susceptible to human mistakes.

Sample Images

Accuracy

The accuracy obtained is 97%

Accuracy

Heatmap

Original Image

Original

After Heatmap

Heatmap

You can download the images with the help of the gi2ds.

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].