Created temporary directory: /tmp/pip-ephem-wheel-cache-codihgm0
Created temporary directory: /tmp/pip-req-tracker-g37mtvg9
Created requirements tracker '/tmp/pip-req-tracker-g37mtvg9'
Created temporary directory: /tmp/pip-wheel-3ukhn142
Looking in indexes: https://pypi.python.org/simple
Collecting density_forest==0.5
  1 location(s) to search for versions of density-forest:
  * https://pypi.python.org/simple/density-forest/
  Getting page https://pypi.python.org/simple/density-forest/
  Analyzing links from page https://pypi.org/simple/density-forest/
    Found link https://files.pythonhosted.org/packages/d4/5c/a84ea0027844b70f6dae8c1a5f26eeaa753b515be7808cef62b31beebde7/density_forest-0.1.tar.gz#sha256=099c951cbd9ceb1b57e46dbd5fbb9baf2e12c0cd7bcfcdafe364359a7de83b59 (from https://pypi.org/simple/density-forest/), version: 0.1
    Found link https://files.pythonhosted.org/packages/af/af/ca09671e88a99f7938659a5c2ef3e86e9ef26a07ca0586c6ea0457a0ee87/density_forest-0.2.tar.gz#sha256=d74aec7437d3688cc190f70cb4eac04c3b2084e2f87a8af501dde53d92bdf96c (from https://pypi.org/simple/density-forest/), version: 0.2
    Found link https://files.pythonhosted.org/packages/72/ff/c6141eea2a4ee843b4c3ee48d45f1afd75559bb0405e736d94f8e4eb0feb/density_forest-0.3.tar.gz#sha256=9c6431df7bcd442cf8629ea1b1a568509f5de5c11bc85f271f0629c1436bd87f (from https://pypi.org/simple/density-forest/), version: 0.3
    Found link https://files.pythonhosted.org/packages/b1/67/65e1bb04984280d14895b2261b68e5220fc94aa0b49c6b0044970a672069/density_forest-0.3.0.1-py3-none-any.whl#sha256=c366783c4ca58acdf1fd8f25b43081a682b62ce4c51238693361c5d3570dbaad (from https://pypi.org/simple/density-forest/), version: 0.3.0.1
    Found link https://files.pythonhosted.org/packages/5d/92/759c3dff2e84713ed2e4f7a83414944e7383183e4191414795c0381d4832/density_forest-0.3.0.1.tar.gz#sha256=08e20881b1db2e548414437fdf8c10e58df4c509df317e308727b8406ef4cb79 (from https://pypi.org/simple/density-forest/), version: 0.3.0.1
    Found link https://files.pythonhosted.org/packages/07/5d/cdf24d62ebf0f757494d32dc340176d99bd2e99226b8a4e6a67221ec2f65/density_forest-0.3.1-py3-none-any.whl#sha256=f75c9db5a8ca26e06eef75cf1ff3354486c381670963741d399c996fa60a9982 (from https://pypi.org/simple/density-forest/), version: 0.3.1
    Found link https://files.pythonhosted.org/packages/60/f7/952ce9000404114cb27cd4b5556930bb851e60d49221d005f1930b734a38/density_forest-0.3.1.tar.gz#sha256=646d54457f11f3da5ebd6643b40eed1fafb0dbf0563587d016dc7f7065db4c86 (from https://pypi.org/simple/density-forest/), version: 0.3.1
    Found link https://files.pythonhosted.org/packages/1d/69/708c8ef11b78c243ebef89b908bb2e254d41a92de90d86a1d470068752ac/density_forest-0.4-py3-none-any.whl#sha256=975fd53efac567763385865e826655df58b72199967c9e76d23cb58c4d1daee4 (from https://pypi.org/simple/density-forest/), version: 0.4
    Found link https://files.pythonhosted.org/packages/1a/17/f618ddcba0d5caa7f7a3951c3e294cff63ff79bd3ac0618d2b3f89a57687/density_forest-0.4.1-py3-none-any.whl#sha256=786c5e552595c31472cf600d124da85d865e3b552a5bca1a200674a9557be053 (from https://pypi.org/simple/density-forest/), version: 0.4.1
    Found link https://files.pythonhosted.org/packages/5b/fd/d0c6eb31b0ae1479a4822e6b3e97c0f3e7466e1393456f8076a2ab0ee5d8/density_forest-0.4.1.tar.gz#sha256=f1b65c28ba089ef72c87b79c1eb655008804f20676c21424e2066fcaba802eb3 (from https://pypi.org/simple/density-forest/), version: 0.4.1
    Found link https://files.pythonhosted.org/packages/12/ed/00593cc4c1f3328bcde56d7221142832bf3402e0edffbe831a5ff71fb59c/density_forest-0.4.2-py3-none-any.whl#sha256=9e084ab51edce8ce6b4d2124a0a07e9c9da0e89db8a57bfe15340e6159b16c87 (from https://pypi.org/simple/density-forest/), version: 0.4.2
    Found link https://files.pythonhosted.org/packages/6a/f0/494777bac5b2ba74a8cf8f5c9706856da409096388ac42d6c1b531db518c/density_forest-0.4.2.tar.gz#sha256=4bc6694df243d3a1ecd847a625c060309dd6f007c4acadf2386be293b51e663a (from https://pypi.org/simple/density-forest/), version: 0.4.2
    Found link https://files.pythonhosted.org/packages/54/36/7db51c19c46cffee5f112f394fc6259dea58da42c63fa704d37c6cd5c8e8/density_forest-0.5.tar.gz#sha256=d2bc7c5a416d9973e8537900fa5ae4a0d60b8250168ac59dd6eed22fca80a4d2 (from https://pypi.org/simple/density-forest/), version: 0.5
  Using version 0.5 (newest of versions: 0.5)
  Created temporary directory: /tmp/pip-unpack-jfmbnm0w
  Downloading https://files.pythonhosted.org/packages/54/36/7db51c19c46cffee5f112f394fc6259dea58da42c63fa704d37c6cd5c8e8/density_forest-0.5.tar.gz
  Downloading from URL https://files.pythonhosted.org/packages/54/36/7db51c19c46cffee5f112f394fc6259dea58da42c63fa704d37c6cd5c8e8/density_forest-0.5.tar.gz#sha256=d2bc7c5a416d9973e8537900fa5ae4a0d60b8250168ac59dd6eed22fca80a4d2 (from https://pypi.org/simple/density-forest/)
  Added density_forest==0.5 from https://files.pythonhosted.org/packages/54/36/7db51c19c46cffee5f112f394fc6259dea58da42c63fa704d37c6cd5c8e8/density_forest-0.5.tar.gz#sha256=d2bc7c5a416d9973e8537900fa5ae4a0d60b8250168ac59dd6eed22fca80a4d2 to build tracker '/tmp/pip-req-tracker-g37mtvg9'
  Running setup.py (path:/tmp/pip-wheel-3ukhn142/density-forest/setup.py) egg_info for package density-forest
    Running command python setup.py egg_info
    # Density Forest
    This library was developed within an EPFL Master Project, Spring Semester 2018.

    GitHub repository: https://github.com/CyrilWendl/SIE-Master


    ## 📖 Usage of the `DensityForest` class:
    #### Fitting a Density Forest
    Suppose you have your training data `X_train` and test data `X_test`, in `[N, D]` with `N` data points in `D` dimensions:

    ```python
    from density_forest.density_forest import DensityForest

    clf_df = DensityForest(**params)  # create new class instance, put hyperparameters here
    clf_df.fit(X_train)               # fit to a training set
    conf = clf_df.predict(X_test)     # get confidence values for test set
    ```
    Hyperparameters are documented in the docstring. To find the optimal hyperparameters, consider the section below.

    #### Finding Hyperparameters
    To find the optimal hyperparameters, use the `ParameterSearch` from `helpers.cross_validator`, which allows CV, and hyperparameter search.

    ```python
    from helpers.cross_validator import ParameterSearch

    # define hyperparameters to test
    tuned_params = [{'max_depth':[2, 3, 4], 'n_trees': [10, 20]}] # optionally add non-default arguments as single-element arrays
    default_params = [{'verbose':0, ...}]  # other default parameters
    # do parameter search
    ps = ParameterSearch(DensityForest, tuned_parameters, X_train, X_train_all, y_true_tr, f_scorer, n_iter=2, verbosity=0, n_jobs=1, default_params=default_params)
    ps.fit()

    # get model with the best parameters, as above
    clf_df = DensityForest(**ps.best_params, **default_params)  # create new class instance with best hyperparameters
    ...  # continue as above
    ```
    Check the docstrings for more detailed documentation af the `ParameterSearch` class.


    ## 🗂 File Structure

    ### 👾 Code
    All libraries for density forests, helper libraries for semantic segmentation and for baselines.
    #### `density_forest/`
    Package for implementation of Decision Trees, Random Forests, Density Trees and Density Forests
    - `create_data.py`: functions for generating labelled and unlabelled data
    - `decision_tree.py`: data structure for decision tree nodes
    - `decision_tree_create.py`: functions for generating decision trees
    - `decision_tree_traverse.py`: functions for traversing a decision tree and predicting labels
    - `density_forest.py`: functions for creating density forests
    - `density_tree.py`: data struture for density tree nodes
    - `density_tree_create.py`: functions for generating a density tree
    - `density_tree_traverse.py`: functions for descending a density tree and retrieving its cluster parameters
    - `helper.py`: various helper functions
    - `random_forests.py`: functions for creating random forests

    #### `helpers/`:
    General helpers library for semantic segmentation
    - `data_augment.py`: custom data augmentation methods applied to both the image and the ground truth
    - `data_loader.py`: PyTorch data loader for Zurich dataset
    - `helpers.py`: functions for importing, cropping, padding images and other related image tranformations
    - `parameter_search.py`: functions for finding optimal hyperparameters for Density Forest, OC-SVM and GMM (explained above)
    - `plots.py`:  Generic plotter functions for labelled and unlabelled 2D and 3D plots, used for t-SNE and PCA plots

    #### `baselines/`:
    Helper functions for confidence estimation baselines MSR, margin, entropy and MC-Dropout

    #### `keras_helpers/`
    Helper functions for Keras
    - `helpers.py`: get activations
    - `callbacks.py`: callbacks to be evaluated after each epoch
    - `unet.py`: UNET model for training of network on Zurich dataset

    ### 🗾 Visualizations
    #### `density_forest/`:
    Visualizations of basic decision tree and density tree
    - `Decision Forest.ipynb`: Decision Trees and Random Forest on randomly generated labelled data
    - `Density Forest.ipynb`: Density Trees on randomly generated unlabelled data

    ## 🎓 Supervisors:
    - Prof. Devis Tuia, University of Wageningen
    - Diego Marcos González, University of Wageningen
    - Prof. François Golay, EPFL

    Cyril Wendl, 2018

    running egg_info
    creating pip-egg-info/density_forest.egg-info
    writing requirements to pip-egg-info/density_forest.egg-info/requires.txt
    writing top-level names to pip-egg-info/density_forest.egg-info/top_level.txt
    writing pip-egg-info/density_forest.egg-info/PKG-INFO
    writing dependency_links to pip-egg-info/density_forest.egg-info/dependency_links.txt
    writing manifest file 'pip-egg-info/density_forest.egg-info/SOURCES.txt'
    reading manifest file 'pip-egg-info/density_forest.egg-info/SOURCES.txt'
    writing manifest file 'pip-egg-info/density_forest.egg-info/SOURCES.txt'
  Source in /tmp/pip-wheel-3ukhn142/density-forest has version 0.5, which satisfies requirement density_forest==0.5 from https://files.pythonhosted.org/packages/54/36/7db51c19c46cffee5f112f394fc6259dea58da42c63fa704d37c6cd5c8e8/density_forest-0.5.tar.gz#sha256=d2bc7c5a416d9973e8537900fa5ae4a0d60b8250168ac59dd6eed22fca80a4d2
  Removed density_forest==0.5 from https://files.pythonhosted.org/packages/54/36/7db51c19c46cffee5f112f394fc6259dea58da42c63fa704d37c6cd5c8e8/density_forest-0.5.tar.gz#sha256=d2bc7c5a416d9973e8537900fa5ae4a0d60b8250168ac59dd6eed22fca80a4d2 from build tracker '/tmp/pip-req-tracker-g37mtvg9'
Building wheels for collected packages: density-forest
  Created temporary directory: /tmp/pip-wheel-a_v717x2
  Running setup.py bdist_wheel for density-forest: started
  Destination directory: /tmp/pip-wheel-a_v717x2
  Running command /usr/bin/python3 -u -c "import setuptools, tokenize;__file__='/tmp/pip-wheel-3ukhn142/density-forest/setup.py';f=getattr(tokenize, 'open', open)(__file__);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, __file__, 'exec'))" bdist_wheel -d /tmp/pip-wheel-a_v717x2
  # Density Forest
  This library was developed within an EPFL Master Project, Spring Semester 2018.

  GitHub repository: https://github.com/CyrilWendl/SIE-Master


  ## 📖 Usage of the `DensityForest` class:
  #### Fitting a Density Forest
  Suppose you have your training data `X_train` and test data `X_test`, in `[N, D]` with `N` data points in `D` dimensions:

  ```python
  from density_forest.density_forest import DensityForest

  clf_df = DensityForest(**params)  # create new class instance, put hyperparameters here
  clf_df.fit(X_train)               # fit to a training set
  conf = clf_df.predict(X_test)     # get confidence values for test set
  ```
  Hyperparameters are documented in the docstring. To find the optimal hyperparameters, consider the section below.

  #### Finding Hyperparameters
  To find the optimal hyperparameters, use the `ParameterSearch` from `helpers.cross_validator`, which allows CV, and hyperparameter search.

  ```python
  from helpers.cross_validator import ParameterSearch

  # define hyperparameters to test
  tuned_params = [{'max_depth':[2, 3, 4], 'n_trees': [10, 20]}] # optionally add non-default arguments as single-element arrays
  default_params = [{'verbose':0, ...}]  # other default parameters
  # do parameter search
  ps = ParameterSearch(DensityForest, tuned_parameters, X_train, X_train_all, y_true_tr, f_scorer, n_iter=2, verbosity=0, n_jobs=1, default_params=default_params)
  ps.fit()

  # get model with the best parameters, as above
  clf_df = DensityForest(**ps.best_params, **default_params)  # create new class instance with best hyperparameters
  ...  # continue as above
  ```
  Check the docstrings for more detailed documentation af the `ParameterSearch` class.


  ## 🗂 File Structure

  ### 👾 Code
  All libraries for density forests, helper libraries for semantic segmentation and for baselines.
  #### `density_forest/`
  Package for implementation of Decision Trees, Random Forests, Density Trees and Density Forests
  - `create_data.py`: functions for generating labelled and unlabelled data
  - `decision_tree.py`: data structure for decision tree nodes
  - `decision_tree_create.py`: functions for generating decision trees
  - `decision_tree_traverse.py`: functions for traversing a decision tree and predicting labels
  - `density_forest.py`: functions for creating density forests
  - `density_tree.py`: data struture for density tree nodes
  - `density_tree_create.py`: functions for generating a density tree
  - `density_tree_traverse.py`: functions for descending a density tree and retrieving its cluster parameters
  - `helper.py`: various helper functions
  - `random_forests.py`: functions for creating random forests

  #### `helpers/`:
  General helpers library for semantic segmentation
  - `data_augment.py`: custom data augmentation methods applied to both the image and the ground truth
  - `data_loader.py`: PyTorch data loader for Zurich dataset
  - `helpers.py`: functions for importing, cropping, padding images and other related image tranformations
  - `parameter_search.py`: functions for finding optimal hyperparameters for Density Forest, OC-SVM and GMM (explained above)
  - `plots.py`:  Generic plotter functions for labelled and unlabelled 2D and 3D plots, used for t-SNE and PCA plots

  #### `baselines/`:
  Helper functions for confidence estimation baselines MSR, margin, entropy and MC-Dropout

  #### `keras_helpers/`
  Helper functions for Keras
  - `helpers.py`: get activations
  - `callbacks.py`: callbacks to be evaluated after each epoch
  - `unet.py`: UNET model for training of network on Zurich dataset

  ### 🗾 Visualizations
  #### `density_forest/`:
  Visualizations of basic decision tree and density tree
  - `Decision Forest.ipynb`: Decision Trees and Random Forest on randomly generated labelled data
  - `Density Forest.ipynb`: Density Trees on randomly generated unlabelled data

  ## 🎓 Supervisors:
  - Prof. Devis Tuia, University of Wageningen
  - Diego Marcos González, University of Wageningen
  - Prof. François Golay, EPFL

  Cyril Wendl, 2018

  running bdist_wheel
  running build
  running build_py
  creating build
  creating build/lib
  creating build/lib/density_forest
  copying density_forest/density_forest.py -> build/lib/density_forest
  copying density_forest/decision_tree.py -> build/lib/density_forest
  copying density_forest/density_tree_create.py -> build/lib/density_forest
  copying density_forest/random_forest.py -> build/lib/density_forest
  copying density_forest/decision_tree_traverse.py -> build/lib/density_forest
  copying density_forest/density_tree.py -> build/lib/density_forest
  copying density_forest/__init__.py -> build/lib/density_forest
  copying density_forest/density_tree_traverse.py -> build/lib/density_forest
  copying density_forest/helpers.py -> build/lib/density_forest
  copying density_forest/decision_tree_create.py -> build/lib/density_forest
  creating build/lib/baselines
  copying baselines/plots.py -> build/lib/baselines
  copying baselines/__init__.py -> build/lib/baselines
  copying baselines/helpers.py -> build/lib/baselines
  creating build/lib/helpers
  copying helpers/create_data.py -> build/lib/helpers
  copying helpers/data_loader.py -> build/lib/helpers
  copying helpers/plots.py -> build/lib/helpers
  copying helpers/parameter_search.py -> build/lib/helpers
  copying helpers/__init__.py -> build/lib/helpers
  copying helpers/helpers.py -> build/lib/helpers
  copying helpers/data_augment.py -> build/lib/helpers
  creating build/lib/keras_helpers
  copying keras_helpers/unet.py -> build/lib/keras_helpers
  copying keras_helpers/callbacks.py -> build/lib/keras_helpers
  copying keras_helpers/__init__.py -> build/lib/keras_helpers
  installing to build/bdist.linux-armv7l/wheel
  running install
  running install_lib
  creating build/bdist.linux-armv7l
  creating build/bdist.linux-armv7l/wheel
  creating build/bdist.linux-armv7l/wheel/keras_helpers
  copying build/lib/keras_helpers/unet.py -> build/bdist.linux-armv7l/wheel/keras_helpers
  copying build/lib/keras_helpers/callbacks.py -> build/bdist.linux-armv7l/wheel/keras_helpers
  copying build/lib/keras_helpers/__init__.py -> build/bdist.linux-armv7l/wheel/keras_helpers
  creating build/bdist.linux-armv7l/wheel/density_forest
  copying build/lib/density_forest/density_forest.py -> build/bdist.linux-armv7l/wheel/density_forest
  copying build/lib/density_forest/decision_tree.py -> build/bdist.linux-armv7l/wheel/density_forest
  copying build/lib/density_forest/density_tree_create.py -> build/bdist.linux-armv7l/wheel/density_forest
  copying build/lib/density_forest/random_forest.py -> build/bdist.linux-armv7l/wheel/density_forest
  copying build/lib/density_forest/decision_tree_traverse.py -> build/bdist.linux-armv7l/wheel/density_forest
  copying build/lib/density_forest/density_tree.py -> build/bdist.linux-armv7l/wheel/density_forest
  copying build/lib/density_forest/__init__.py -> build/bdist.linux-armv7l/wheel/density_forest
  copying build/lib/density_forest/density_tree_traverse.py -> build/bdist.linux-armv7l/wheel/density_forest
  copying build/lib/density_forest/helpers.py -> build/bdist.linux-armv7l/wheel/density_forest
  copying build/lib/density_forest/decision_tree_create.py -> build/bdist.linux-armv7l/wheel/density_forest
  creating build/bdist.linux-armv7l/wheel/baselines
  copying build/lib/baselines/plots.py -> build/bdist.linux-armv7l/wheel/baselines
  copying build/lib/baselines/__init__.py -> build/bdist.linux-armv7l/wheel/baselines
  copying build/lib/baselines/helpers.py -> build/bdist.linux-armv7l/wheel/baselines
  creating build/bdist.linux-armv7l/wheel/helpers
  copying build/lib/helpers/create_data.py -> build/bdist.linux-armv7l/wheel/helpers
  copying build/lib/helpers/data_loader.py -> build/bdist.linux-armv7l/wheel/helpers
  copying build/lib/helpers/plots.py -> build/bdist.linux-armv7l/wheel/helpers
  copying build/lib/helpers/parameter_search.py -> build/bdist.linux-armv7l/wheel/helpers
  copying build/lib/helpers/__init__.py -> build/bdist.linux-armv7l/wheel/helpers
  copying build/lib/helpers/helpers.py -> build/bdist.linux-armv7l/wheel/helpers
  copying build/lib/helpers/data_augment.py -> build/bdist.linux-armv7l/wheel/helpers
  running install_egg_info
  running egg_info
  writing dependency_links to density_forest.egg-info/dependency_links.txt
  writing top-level names to density_forest.egg-info/top_level.txt
  writing requirements to density_forest.egg-info/requires.txt
  writing density_forest.egg-info/PKG-INFO
  reading manifest file 'density_forest.egg-info/SOURCES.txt'
  writing manifest file 'density_forest.egg-info/SOURCES.txt'
  Copying density_forest.egg-info to build/bdist.linux-armv7l/wheel/density_forest-0.5-py3.4.egg-info
  running install_scripts
  creating build/bdist.linux-armv7l/wheel/density_forest-0.5.dist-info/WHEEL
  creating '/tmp/pip-wheel-a_v717x2/density_forest-0.5-py3-none-any.whl' and adding '.' to it
  adding 'baselines/__init__.py'
  adding 'baselines/helpers.py'
  adding 'baselines/plots.py'
  adding 'density_forest/__init__.py'
  adding 'density_forest/decision_tree.py'
  adding 'density_forest/decision_tree_create.py'
  adding 'density_forest/decision_tree_traverse.py'
  adding 'density_forest/density_forest.py'
  adding 'density_forest/density_tree.py'
  adding 'density_forest/density_tree_create.py'
  adding 'density_forest/density_tree_traverse.py'
  adding 'density_forest/helpers.py'
  adding 'density_forest/random_forest.py'
  adding 'helpers/__init__.py'
  adding 'helpers/create_data.py'
  adding 'helpers/data_augment.py'
  adding 'helpers/data_loader.py'
  adding 'helpers/helpers.py'
  adding 'helpers/parameter_search.py'
  adding 'helpers/plots.py'
  adding 'keras_helpers/__init__.py'
  adding 'keras_helpers/callbacks.py'
  adding 'keras_helpers/unet.py'
  adding 'density_forest-0.5.dist-info/top_level.txt'
  adding 'density_forest-0.5.dist-info/WHEEL'
  adding 'density_forest-0.5.dist-info/METADATA'
  adding 'density_forest-0.5.dist-info/RECORD'
  removing build/bdist.linux-armv7l/wheel
  Running setup.py bdist_wheel for density-forest: finished with status 'done'
  Stored in directory: /tmp/tmp405ew_o9
Successfully built density-forest
Cleaning up...
  Removing source in /tmp/pip-wheel-3ukhn142/density-forest
Removed build tracker '/tmp/pip-req-tracker-g37mtvg9'