diff --git a/README.md b/README.md index 7a6d4d59a3d6a15b40486b8c38a91c3dc5c992e4..51f1f617034cc5e80600f834f053c9315ee1869e 100644 --- a/README.md +++ b/README.md @@ -1,35 +1,35 @@ -# Compairing performances across multiple Keras models by training them and applying transfer learning. +# Compairing performances across multiple Keras models by training them and applying transfer learning on 3D images datasets. -Author: alexcla99 +Author: aclausse Version: 2.0.0 ### Folder content: ``` -+-+- ft_data/ # The folder containing the fine tuning dataset -| +--- coma/ # The folder containing coma MRIs -| +--- control/ # The folder containing control MRIs ++-+- multi_train # A folder containing scripts to test a model with cross configurations +| +--- dataset.py # An override of the original dataset.py file +| +--- fine_tune_v2.py # A new version of fine_tune.py file +| +--- train_all.py # A script to run both training and fine tuning with multiple configurations +| +--- train_all_balance.py # The same script as above with balancing the train dataset +| +--- train_v2.py # A new version of train.py file | -+-+- train_data/ # The folder containing the train dataset -| +--- abnormal/ # The folder containing abnormal MRIs -| +--- control/ # The folder containing control MRIs ++-+- models/ # The folder containing available models +| +--- LeNet17.py # The LeNet17 model | -+-+- models/ # The folder containing available models -| +--- LeNet17.py # The LeNet17 model -| -+--- results/ # The folder containing the train, transfer learning and tests results -+--- __init__.py # An empty file to make this directory being a Python library -+--- dataset.py # The dataset loader -+--- fine_tune.py # A script to apply fine tuning on a model -+--- README.md # This file -+--- requirements.txt # The Python libraries to be installed in order to run the project -+--- Results.ipynb # The obtained results and a description on how the model has been fine-tuned -+--- settings.json # The settings of the model and the train phase -+--- test_trained_model.py # A script to test a trained model -+--- test_fine_tuned_model.py # A script to test a fine_tuned model -+--- tf_config.py # A script to configure TensorFlow -+--- train.py # A script to train from scratch a model -+--- utils.py # Some utils ++--- results/ # The folder containing the train, transfer learning and tests results ++--- train_data/ # The folder containing the dataset for training from scratch ++--- ft_data/ # The folder containing the dataset for fine tuning ++--- __init__.py # An empty file to make this directory being a Python library ++--- dataset.py # The dataset loader ++--- fine_tune.py # A script to apply fine tuning on a model ++--- README.md # This file ++--- requirements.txt # The Python libraries to be installed in order to run the project ++--- settings.json # The settings of the model and the train phase ++--- test_trained_model.py # A script to test a trained model ++--- test_fine_tuned_model.py # A script to test a fine tuned model ++--- tf_config.py # A script to configure TensorFlow ++--- train.py # A script to train from scratch a model ++--- utils.py # Some utils ``` ### Usage: @@ -70,12 +70,9 @@ python3 test_fine_tuned_model.py <model:str> # Example: python3 test_fine_tuned_model.py LeNet17 ``` -### Obtained results: - -See the `Results.ipynb` file. - ### Many thanks to: -3D images classification: https://keras.io/examples/vision/3D_image_classification/ +[1] H. Zunair., "3D images classification from CT scans.", [keras.io](https://keras.io/examples/vision/3D_image_classification/), 2020. +[2] Zunair et al., "Uniformizing Techniques to Process CT scans with 3D CNNs for Tuberculosis Prediction", [arXiv](https://arxiv.org/pdf/2007.13224.pdf), 2020. -LeNet17: https://arxiv.org/pdf/2007.13224.pdf \ No newline at end of file +License: [Apache 2.0](http://www.apache.org/licenses/LICENSE-2.0).