features_analysis.py

# Copyright CNRS/Inria/UNS
# Contributor(s): Eric Debreuve (since 2019), Morgane Nadal (2020)
#
# eric.debreuve@cnrs.fr
#
# This software is governed by the CeCILL  license under French law and
# abiding by the rules of distribution of free software.  You can  use,
# modify and/ or redistribute the software under the terms of the CeCILL
# license as circulated by CEA, CNRS and INRIA at the following URL
# "http://www.cecill.info".
#
# As a counterpart to the access to the source code and  rights to copy,
# modify and redistribute granted by the license, users are provided only
# with a limited warranty  and the software's author,  the holder of the
# economic rights,  and the successive licensors  have only  limited
# liability.
#
# In this respect, the user's attention is drawn to the risks associated
# with loading,  using,  modifying and/or developing or reproducing the
# software by the user in light of its specific status of free software,
# that may mean  that it is complicated to manipulate,  and  that  also
# therefore means  that it is reserved for developers  and  experienced
# professionals having in-depth computer knowledge. Users are therefore
# encouraged to load and test the software's suitability as regards their
# requirements in conditions enabling the security of their systems and/or
# data to be ensured and,  more generally, to use and operate it in the
# same conditions as regards security.
#
# The fact that you are presently reading this means that you have had
# knowledge of the CeCILL license and that you accept its terms.

import pandas as pd_
import numpy as np_
import matplotlib.pyplot as pl_
from sklearn.cluster import KMeans
from sklearn.decomposition import PCA
from sklearn.preprocessing import StandardScaler
import scipy as si_
import seaborn as sb_
import os
import glob
from typing import List


def KmeansOnDF(df: pd_.DataFrame(),
               nb_clusters: tuple,
               representation: bool = False,
               labeled_somas=None,
               elbow: bool = False,
               intracluster_var: bool = True,
               ) -> KMeans:
    '''
    Perform kmeans on the pandas dataframe. Can find the best number of cluster with elbow method,
    find the intracluster variance, and represent the result on the initial images.
    Returns kmeans.
    '''
    # Data standardization
    scaler = StandardScaler()
    scaler.fit(df)
    stand_df = scaler.transform(df)

    # Best number of clusters using Elbow method
    if elbow:
        wcss = []  # within cluster sum of errors(wcss)
        for i in range(1, 24):
            kmeans = KMeans(n_clusters=i, init='k-means++', max_iter=300, n_init=10, random_state=0)
            kmeans.fit(stand_df)
            wcss.append(kmeans.inertia_)
        pl_.plot(range(1, 24), wcss)
        pl_.plot(range(1, 24), wcss, 'bo')
        pl_.title('Elbow Method')
        pl_.xlabel('Number of clusters')
        pl_.ylabel('WCSS')
        pl_.show(block=True)
        pl_.close()

    # Kmeans with x clusters
    for nb_cluster in nb_clusters:
        kmeans = KMeans(n_clusters=nb_cluster, init='k-means++', max_iter=300, n_init=10, random_state=0)
        pred_y = kmeans.fit_predict(stand_df)

        # Intracluster variance
        if intracluster_var:
            var = IntraClusterVariance(df, kmeans, nb_cluster)

        # Representation on the image
        if representation:
            RepresentationOnImages(labeled_somas, kmeans, nb_cluster)

    return kmeans


def IntraClusterVariance(df: pd_.DataFrame(), kmeans: KMeans(), nb_cluster: int) -> list:
    '''
    Return the intracluster variance of a given cluster found by kmeans.
    '''
    var = []
    for cluster in range(nb_cluster):
        soma_cluster = [indx + 1 for indx, value in enumerate(kmeans.labels_) if value == cluster]
        mean_cluster = np_.average([df.loc[f"soma {row}", :] for row in soma_cluster], axis=0)
        variance = sum([np_.linalg.norm(df.loc[f"soma {row}", :] - mean_cluster) ** 2 for row in soma_cluster]) / len(
            soma_cluster)
        var.append(variance)

    print(f"Intracluster variance for {nb_cluster} clusters :", var)

    return var


def RepresentationOnImages(labeled_somas, kmeans, nb_cluster):
    '''
    Represent the result of kmeans on labeled image. IN DVPT. Only available for a kmean intra-image.
    '''
    clustered_somas = labeled_somas.copy()
    clustered_somas = np_.amax(clustered_somas, axis=0)
    for indx, value in enumerate(kmeans.labels_):
        for indx_axe, axe in enumerate(clustered_somas):
            for indx_pixel, pixel in enumerate(axe):
                if pixel == indx + 1:
                    clustered_somas[indx_axe][indx_pixel] = value + 1
    pl_.imshow(clustered_somas, cmap="tab20")
    pl_.title(f"n cluster = {nb_cluster}")
    pl_.show(block=True)
    pl_.close()


def FeaturesStatistics(df):
    '''
    Return the statistics allowing the user to choose the most relevant features to feed ML algorithms for ex.
    '''
    # Overview of the basic stats on each columns of df
    df.describe()

    # Overview of features distribution and correlation
    sb_.pairplot(df, kind='reg')
    sb_.pairplot(df, kind='reg', hue='Condition')

    # Statistics for each features
    for column in df.columns:
        print(column)
        hist = df[column].hist(bins=20)
        pl_.title(f"{column}")
        pl_.savefig(f"D:\\MorganeNadal\\M2 report\\kmeans24\\feat_distrib_{column}.png")
        pl_.close()


def PCAOnDF(df: pd_.DataFrame(),
            target: str,
            targets: List[str],
            colors: List[str],
            save_name: str = None,
            title: str = ""
            ) -> list:
    '''
    Perform 2D PCA on the CHO-DIO dataframe.
    Print ratio variance and plot the PCA.
    '''
    # Separating the features from their conditions and durations
    # all_target = df.loc[:, [target]].values
    df_all = df.drop([target], axis=1)

    # Standardize the data
    scaler = StandardScaler()
    scaler.fit(df_all)
    stand_df = scaler.transform(df_all)

    # Create the PCA and fit the data
    pca = PCA(n_components=2)
    principal_components = pca.fit_transform(stand_df)
    print(f"PCA explained variance ratio ({save_name}): ", pca.explained_variance_ratio_)

    principal_df = pd_.DataFrame(data=principal_components, columns=['principal component 1', 'principal component 2'])

    # Give the final df containing the principal component and their condition
    final_df = pd_.concat([principal_df, df[target]], axis=1)

    # Plot
    fig = pl_.figure(figsize=(8, 8))
    ax = fig.add_subplot(1, 1, 1)
    ax.set_xlabel('Principal Component 1', fontsize=15)
    ax.set_ylabel('Principal Component 2', fontsize=15)
    ax.set_title(f'2 component PCA{title}', fontsize=20)
    for tgt, color in zip(targets, colors):
        idx = final_df[target] == tgt
        ax.scatter(final_df.loc[idx, 'principal component 1']
                   , final_df.loc[idx, 'principal component 2']
                   , c=color
                   , s=30)
    ax.legend(targets)
    ax.grid()
    if save_name is not None:
        pl_.savefig(f"D:\\MorganeNadal\\2_RESULTS\\Results_wo_erosion\\Features_analysis\\PCA_{save_name}.png")

    return pca.explained_variance_ratio_


if __name__ == "__main__":
    #
    # os.chdir("D:\\MorganeNadal\\2_RESULTS\\Results_wo_erosion")

    ## If need to concatenate files:
    # all_filenames = [i for i in glob.glob('*.{}'.format("csv"))]
    # print(all_filenames)
    # df = pd_.concat([pd_.read_csv(f, index_col=0) for f in all_filenames])
    # df.to_csv(".\combined_features.csv")

    ## If use labeled somas:
    # labeled_somas = np_.load("D:\\MorganeNadal\\Results\\labeled_somas.npy")
    # df = pd_.read_csv(".\combined_features.csv", index_col=0)

    df0 = pd_.read_csv("D:\\MorganeNadal\\2_RESULTS\\Results_wo_erosion\\all_features.csv",
                      # index_col=0,
                      )
    df = df0.drop(["Unnamed: 0"], axis=1)

    # Statistical analysis

    # For the moment drop the columns with non scalar values, and un-useful values
    # - TO BE CHANGED (use distance metrics such as bhattacharyya coef, etc)
    df = df.drop(["soma uid",
                  "spherical_angles_eva", "spherical_angles_evb",
                  "hist_lengths", "hist_lengths_P", "hist_lengths_S",
                  "hist_curvature", "hist_curvature_P", "hist_curvature_S"],
                 axis=1)
    df = df.dropna(axis=0, how="any")

    # KMeansIntraImage(df, nb_clusters=(2,))
    # FeatureDistribution(df)

    # -- PCA with all the features
    # Between the two conditions, regardless the duration of experiment (2 conditions, all durations)
    df_all = df.drop(["duration"], axis=1)
    all_pca = PCAOnDF(df_all, target="condition", targets=["CHO", "DIO"], colors=["b", "r"], save_name="all_features")

    # Between the two conditions, for each duration (2 conditions, 3 durations)
    groupby_duration = df.groupby("duration")
    duration_pca = []
    for duration, values in groupby_duration:
        # print(duration, values.shape)
        # groupby_condition = values.groupby("condition")
        # for cond, val in groupby_condition:
        #     print(cond, val.shape)
        ## duration: str, values: pd_.DataFrame()
        duration_df = values.drop(["duration"], axis=1)
        pca = PCAOnDF(duration_df,
                      target="condition",
                      targets=["CHO", "DIO"],
                      colors=["b", "r"],
                      save_name=f"{duration}_features",
                      title=f" - {duration} Sample")
        duration_pca.append(pca)

    ## -- K-means with all the features (2 conditions)

    ## Between the two conditions, regardless the duration of experiment (2 conditions, all durations)
    # kmeans = fa_.KmeansOnDF(concatenated_features_df, nb_clusters=(2,), elbow=True, intracluster_var=True)

    ## Between the two conditions, for each duration (2 conditions, 3 durations)
    # for duration, values in groupby_duration:
    # kmeans = fa_.KmeansOnDF(values, nb_clusters=(2,), elbow=True, intracluster_var=True)

    ## -- Select Discriminant features by statistical analysis
    # TODO filtered_df = SelectFeatures(concatenated_features_df)

    ## -- PCA with all the features with the cluster as label

    ## Between the two conditions, regardless the duration of experiment (2 conditions, all durations)
    # TODO pca = fa_.PCAOnDF(concatenated_features_df)

    ## Between the two conditions, for each duration (2 conditions, 3 durations)
    # for duration, values in groupby_duration:
    # pca = fa_.PCAOnDF(values)

    ## -- Select Discriminant features by statistical analysis
    # TODO filtered_df = SelectFeatures(concatenated_features_df)

    ## -- PCA with selected features

    ## Between the two conditions, regardless the duration of experiment (2 conditions, all durations)
    # TODO pca = fa_.PCAOnDF(filtered_df)

    ## Between the two conditions, for each duration (2 conditions, 3 durations)
    # filtered_groupby_duration = filtered_df.groupby("Duration")
    # for duration, values in filtered_groupby_duration:
    # pca = fa_.PCAOnDF(values)

    ## -- K-means with selected features

    ## Between the two conditions, regardless the duration of experiment (2 conditions, all durations)
    # filtered_kmeans = fa_.KmeansOnDF(filtered_df, nb_clusters=(2,), elbow=True, intracluster_var=True)

    ## Between the two conditions, for each duration (2 conditions, 3 durations)
    # for duration, values in filtered_groupby_duration:
    # filtered_kmeans = fa_.KmeansOnDF(values, nb_clusters=(2,), elbow=True, intracluster_var=True)