uptake/autofocus

Issues

• Automate container push with Travis

  #111 opened 5 years ago by gsganden
  2

• Use Megadetector

  #129 opened 5 years ago by gsganden
  3

• Add a contribution guide

  #119 opened 5 years ago by OscarAblinger
  6

• Combine 2016-2017 and 2012-2014 datasets

  #120 opened 5 years ago by gsganden
  15

• More clarity in autofocus output

  #117 opened 5 years ago by Manusreekumar
  9

• Extract features from empty images and treat presence of animals as an anomaly detection problem

  #37 opened 6 years ago by gsganden
  4

• Identify potential animals and crop to them before applying classifier

  #38 opened 6 years ago by gsganden
  2

• Use `mimetypes` to get a list of image file extensions that the OS recognizes

  #112 opened 5 years ago by gsganden
  0

• Clean up app code

  #104 opened 5 years ago by gsganden
  3

• Add a webpage interface for interacting with Autofocus

  #116 opened 5 years ago by Manusreekumar
  3

• Jupyter notebook for un cluttering README

  #118 opened 5 years ago by Manusreekumar
  0

• Host Docker Container on Uptake

  #115 opened 5 years ago by bburns632
  0

• Try pretraining with North American Camera Trap Images (NACTI) dataset

  #92 opened 6 years ago by gsganden
  1

• Do more augmentations

  #58 opened 6 years ago by gsganden
  2

• Run tests on Travis

  #110 opened 5 years ago by gsganden
  0

• autofocus won't predict with JPG (all caps)

  #105 opened 5 years ago by mfidino
  2

• Add JPG as a photo type

  #66 opened 5 years ago by mfidino
  1

• Publish prediction service as a container on dockerhub

  #78 opened 5 years ago by gsganden
  0

• Version floor on pydocstyle is not ideal

  #101 opened 5 years ago by gsganden
  0

• Make Lincoln Park Zoo 2012-2014 dataset available

  #90 opened 6 years ago by gsganden
  0

• Make Lincoln Park Zoo 2018 dataset available

  #96 opened 6 years ago by gsganden
  0

• Make new Lincoln Park Zoo dataset available

  #97 opened 6 years ago by gsganden
  0

• add CODEOWNERS file

  #91 opened 6 years ago by jameslamb
  2

• Use exif times

  #93 opened 6 years ago by gsganden
  0

• Add pycodestyle step on Travis

  #70 opened 6 years ago by jameslamb
  1

• Add Travis CI badge

  #69 opened 6 years ago by jameslamb
  0

• Make a CONTRIBUTING.md and/or PR checklist template

  #57 opened 6 years ago by gsganden
  0

• Add labels to issues

  #50 opened 6 years ago by gsganden
  1

• Provide simple, clean notebook to help people get started working with the data

  #56 opened 6 years ago by smasis001
  5

• Organize issues into milestones

  #68 opened 6 years ago by jameslamb
  1

• Incorporate a background subtraction step

  #36 opened 6 years ago by gsganden
  1

• Try pretraining on Zamba dataset

  #39 opened 6 years ago by gsganden
  0

• Try converting convolutional kernels of pretrained network to grayscale and applying to night images

  #40 opened 6 years ago by gsganden
  0

• Support using LIME (or similar) to see to what extent the model is paying attention to animals vs. other cues such as night/day and the background.

  #41 opened 6 years ago by gsganden
  0

• Create centralized location for mlflow logs

  #42 opened 6 years ago by gsganden
  1

• Try off-the-shelf OpenCV pedestrian detector

  #47 opened 6 years ago by gsganden
  0

• Look at clusters in feature space

  #49 opened 6 years ago by gsganden
  0

• Address challenges raised by object of interest occupying a small part of the image

  #59 opened 6 years ago by gsganden
  0

• Investigate hierarchical classification methods and metrics

  #61 opened 6 years ago by gsganden
  0

• Extracting seasons from the images labels

  #67 opened 6 years ago by Manusreekumar
  0

• get_raw.py and get_processed.py could just be AWSCLI commands

  #79 opened 6 years ago by gsganden
  1

• Error in preprocess_images.py if exif data for DateTime is not present for images

  #73 opened 6 years ago by Manusreekumar
  0

• Build up image sizes

  #43 opened 6 years ago by gsganden
  0

• Try fastai library

  #46 opened 6 years ago by gsganden
  0

• Support multitask learning

  #48 opened 6 years ago by gsganden
  0

• Data not available

  #51 opened 6 years ago by gsganden
  0

• Add test coverage to CI

  #71 opened 6 years ago by jameslamb
  1

• tests are broken

  #72 opened 6 years ago by jameslamb
  1

• Windows CI not set up

  #63 opened 6 years ago by jameslamb
  0

• retrain.py follows Google convention of two-space indents whereas rest of repo uses four-space indents

  #44 opened 6 years ago by gsganden
  1