commented out calls to matplotlib.plt to prevent cluster errors

setup.py

@@ -9,7 +9,6 @@ setup(
     maintainer_email='andrei-claudiu.roibu@dtc.ox.ac.uk',
     install_requires=[
         'pip',
-        'matplotlib',
         'numpy',
         'pandas',
         'torch==1.4',

utils/DSbiobank.py

@@ -17,7 +17,6 @@ Usage:
 import numpy as np
 from fsl.data.image import Image
 from fsl.utils.image.resample import resampleToPixdims
-import matplotlib.pyplot as plt
 from preprocessor import directory_reader, regression_weight_calculator
 from tempfile import TemporaryFile
 from datetime import datetime

utils/data_evaluation_utils.py

@@ -21,7 +21,6 @@ import torch
 import logging
 import utils.data_utils as data_utils
 from utils.common_utils import create_folder
-import matplotlib.pyplot as plt
 import pandas as pd
 from fsl.data.image import Image
 from fsl.utils.image.roi import roi
@@ -511,39 +510,3 @@ def _pearson_correlation(volume, target):
         np.sum(np.power(np.subtract(volume, volume.mean()), 2)), np.sum(np.power(np.subtract(target, target.mean()), 2))))
 
     return r
-
-
-def _generate_correlation_matrix(correlation_matrix, title, prediction_output_path, normalize=False):
-    """Visual correlation matrix generator
-
-    This function generates a visual representation of the correlation matrix
-
-    Args:
-        correaltion_matrix (np.array): Array containing the correlation matrix
-        title (str): Title of the correlation matrix figure (also the experiment name)
-        normalize (bool): Flag indicating if the values of the correlation matrix should be normalized
-        prediction_output_path (str): Output prediction path
-    """
-
-    # THIS FUNCTION NEEDS TO BE VERIFIED!
-
-    if normalize:
-        correlation_matrix = correlation_matrix.astype(
-            'float') / correlation_matrix.sum(axis=1)[:, np.newaxis]
-
-    plt.imshow(correlation_matrix, interpolation='nearest', cmap=plt.cm.Blues)
-    plt.title(title)
-    plt.colorbar()
-    tick_marks = np.arange(len(correlation_matrix))
-    plt.xticks(tick_marks, correlation_matrix)
-    plt.yticks(tick_marks, correlation_matrix)
-    threshold = correlation_matrix.max() / 2.0
-    for i, j in itertools.product(range(correlation_matrix.shape[0]), range(correlation_matrix.shape[1])):
-        plt.text(j, i, format(correlation_matrix[i, j], '.4f'), horizontalalignment='center',
-                 color="white" if correlation_matrix[i, j] > threshold else "black")
-
-    plt.tight_layout()
-    plt.ylabel('Predicted')
-    plt.xlabel('Targets')
-
-    plt.savefig(prediction_output_path + '/' + title + '.png')

utils/data_logging_utils.py

@@ -13,8 +13,6 @@ Usage:
 """
 
 import os
-import matplotlib
-import matplotlib.pyplot as plt
 import shutil
 import logging
 import numpy as np
@@ -24,11 +22,8 @@ import torch
 # More here: https://tensorboardx.readthedocs.io/en/latest/tensorboard.html
 
 from tensorboardX import SummaryWriter
-
 import utils.data_evaluation_utils as evaluation
 
-plt.axis('scaled')
-
 
 class LogWriter():
 
@@ -46,7 +41,7 @@ class LogWriter():
         confusion_matrix_cmap (class): Colour Map to be used for the Conusion Matrix
     """
 
-    def __init__(self, number_of_classes, logs_directory, experiment_name, use_last_checkpoint=False, labels=None, confusion_matrix_cmap=plt.cm.Blues):
+    def __init__(self, number_of_classes, logs_directory, experiment_name, use_last_checkpoint=False, labels=None):
 
         self.number_of_classes = number_of_classes
         training_logs_directory = os.path.join(
@@ -66,8 +61,6 @@ class LogWriter():
             'validation': SummaryWriter(logdir=validation_logs_directory)
         }
 
-        self.confusion_matrix_color_map = confusion_matrix_cmap
-
         self.current_iteration = 1
 
         self.labels = ['rsfMRI']