How to make a first (honorable) submission to The Natue Conservancy Fisheries Monitoring challenge.
We use a deep learning pretrained models (already trained on Imagenet) and extract deep Features from it. Each images is now summarize by a vector of size 2048. This part is done by here. You have to download the image from the Kaggle website to do it (Don't do it by yourself it takes a while...) The pretrained inception v3 checkpoint can be download here.
In your virtual machine you have 4 .txt in this repository with the labels/features/image_name of the Fish dataset:
- fish_features.txt: inception_v3 features of the training set
- fish_labels.txt: labels of the training set
- fish_features_test.txt: inception_v3 features of the test set
- pic_names_test.py: name of the pictures of the test set The .txt files can be download here if you are not working with the VM. Don't forget to add them in this directory on your laptop.
Complete the classif_fish_exo.py. You will create a classification from the dee features. Feel free to build the network you want. Running this code will create a pred.csv file in this directory which corresponds to the submission file required by Kaggle.
A simple softmax regression with 10 epochs leads to a score of ~1.14 (rank: 300/1100). Improve it and submit the pred.csv in Kaggle! ;)