CIFAR ISIC

edit the train_dir flag in isic_train.py
edit the data_dir flag in cifar10.py
edit the dir flags in isic_eval.py to reflect their locations
Place the binaries generated with isic_cnn into the data_dir
ensure the filenames within the functions inputs and distorted inputs are the same as the names of the binaries generated with isic_cnn
ensure that the xrange functions reflect the numbers of binaries generated
run isic_train.py (this will begin the training process. It will save every 1000 steps within your indicated train_dir)
if you want to visualize training run tensorboard pointing at the train_dir
When sufficient steps are complete (24k to recreate the experiment) terminate the process of training
before evaluation change batch_size to 10 in cifar10.py
To test run isic_eval.py
this will output two csv files for analysis. (labels and success)

Name		Name	Last commit message	Last commit date
Latest commit History 59 Commits
.gitignore		.gitignore
BUILD		BUILD
README.md		README.md
Summary.ods		Summary.ods
Summary.xlsx		Summary.xlsx
Summary_60k.ods		Summary_60k.ods
__init__.py		__init__.py
bin_test.py		bin_test.py
cifar10.py		cifar10.py
cifar10_eval.py		cifar10_eval.py
cifar10_input.py		cifar10_input.py
cifar10_input_test.py		cifar10_input_test.py
cifar10_multi_gpu_train.py		cifar10_multi_gpu_train.py
cifar10_train.py		cifar10_train.py
isic_eval.py		isic_eval.py
isic_input.py		isic_input.py
isic_train.py		isic_train.py

charlieLehman/cifar_isic