Skip to content

bernardo-cruz/BigdataBootcampChallenge

Repository files navigation

Dear reader,

In the following folders you will find three different models under the notebook Final_Challenge_Hubacher_Freire_Jenni.ipynb. Two are made with the transfer learning method and one with the CNN method.

You will find the pictures in the raw_data folder.

The models, which you can view, can be found in the folder "models". In this folder, the model transfer_learning_trainables_true.h5 is the most promising. But the model resnet_model-2.h5 also achieves good accuracy. In the CNN method, the 2_conv_128_nodes-Batchnorm-Dropout-20220219112148.h5 model was the best, although this lags far behind the other two.

See the notebook comparison.ipynb for a comparison between the different models created during the elaboration. We hope you enjoy looking through our project.

With best regards

Bernardo, Guido and Lars

About

Repository for BigdataBootcampChallenge group work

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published