Skip to content

drivendataorg/naive-bees-classifier

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation



Banner Image

Naive Bees Classifier

Goal of the Competition

Metis wants to know: using images from BeeSpotter can you identify a bee as a honey bee or a bumble bee? These bees have different behaviors and appearances, but given the variety of backgrounds, positions, and image resolutions it can be a challenge for machines to tell them apart.

Wild bees are important pollinators and the spread of colony collapse disorder has only made their role more critical. Right now it takes a lot of time and effort for researchers to gather data on wild bees. Using data submitted by citizen scientists, Bee Spotter is making this process easier. However, they still require that experts examine and identify the bee in each image. You're goal is to build an algorithm to help. The starting point is to determine the genus—Apis (honey bee) or Bombus (bumble bee)—based on photographs of the insects.

What's in this Repository

This repository contains code volunteered from leading competitors in the Naive Bees Classifier on DrivenData. Code for all winning solutions are open source under the MIT License.

Winning code for other DrivenData competitions is available in the competition-winners repository.

Winning Submissions

Place Team or User Public Score Private Score Summary of Model
1 E.A. 0.9951 0.9956 We fine-tuned a GoogleNet model using caffe and achieved great results. Fine-tuning using caffe is an easy task. For each of the inception outputs in GoogleNet, we extracted the features and passed it through logistic regression model.
2 L.V.S. 0.9926 0.99323 To get higher accuracy, I decided to fine-tune a model pre-trained on ImageNet data.
3 loweew 0.9924 0.99316 I used the pre-trained googlenet model provided by caffe as a starting point and fine-tuned on the data sets. Using the last recorded accuracy for each training run, I took the top 75% of models (12 of 16) by accuracy on the validation set. These models were used to predict on the test set and predictions were averaged with equal weighting.

Winner's Interview: "Building the best naïve bees classifier!"

Benchmark Blog Post: "Bumble bee or honey bee?"

About

Competition results for the Naive Bees Classifier competition https://www.drivendata.org/competitions/8/

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •