Petals to the Metal - Flower Classification on TPU is a Getting Started competition designed to help users learn how to use Tensor Processing Units (TPUs) on Kaggle. TPUs are powerful hardware accelerators specialized in deep learning tasks, originally developed by Google for processing large image databases.
The Challenge Participants are challenged to build a machine learning model that identifies the type of flowers in a dataset of images. While there are over 400,000 different types of flowers in the natural world, this competition focuses on just over 100 types for simplicity.
Who Can Participate This competition is created for people with little to no machine learning background. It serves as a non-competitive way to get familiar with Kaggle's platform, learn basic machine learning concepts, and start meeting people in the community. TPU quotas are available at no cost to users.
Format and Submission Requirements
This is a Code Competition where submissions must be made via an output from a Kaggle Notebook or Script. Key requirements include:
* Notebook run-time per session is capped at 3 hours.
* CPU, GPU, or TPU are all allowed.
* The submission file must contain predictions for the test set following the format of the sample_submission.csv file.
* Your code will not be re-run, as there is no completely hidden test set.
Evaluation Submissions are evaluated on macro F1 score. In "macro" F1, a separate F1 score is calculated for each class/label and then averaged.
Prizes Getting Started competitions have no cash prize, medals, or points associated with leaderboard ranking. The rankings are purely for the benefit of users to learn and see how their approach is improving.