How to use the pydot.find_graphviz function in pydot

def createXdot(dotdata,prog='dot'):
    # The following code is from the pydot module written by Ero Carrera
    progs = pydot.find_graphviz()
    
    #prog = 'dot'
    if progs is None:
        return None
    if not progs.has_key(prog):
        log.warning('Invalid prog=%s',prog)
        # Program not found ?!?!
        return None

    tmp_fd, tmp_name = tempfile.mkstemp()
    os.close(tmp_fd)
    f = open(tmp_name,'w')
    f.write(dotdata)
    f.close()
    format = 'xdot'
    cmd = progs[prog]+' -T'+format+' '+tmp_name

'''Using an already trained model, run predictions on unknown images.'''
import json
import tqdm
import numpy as np
import matplotlib.pyplot as plt
import os

from train import DataClassifier
import pydot
pydot.find_graphviz = lambda: True
from keras.models import load_model

def get_model_predictions(model, data_classifier):

    test_character_name_to_npz_path = {
        **data_classifier.partition_to_character_name_to_npz_paths['test'],
        **data_classifier.partition_to_character_name_to_npz_paths['validation']
    }

    character_to_predictions = {}

    flattened = [(character_name, npz_path) for character_name, npz_paths in test_character_name_to_npz_path.items() for npz_path in npz_paths]
    for character_name, npz_path in tqdm.tqdm(flattened):
        npz_name = os.path.basename(npz_path)
        pixels = np.load(npz_path)['pixels']
        predicted_labels = model.predict(np.array([pixels]), batch_size=1)

def _programs_default(self):
        """ Trait initaliser.
        """
        progs = find_graphviz()
        if progs is None:
            logger.warning("GraphViz's executables not found")
            return {}
        else:
            return progs

def __check_graphviz():    
    import pydot
    if pydot.find_graphviz() is None:
        return False
    
    return True

model_base = Model(input, output)

    output = BatchNormalization()(output)
    output = Dropout(0.5)(output)
    output = Dense(128, activation='relu')(output)
    output = BatchNormalization()(output)
    output = Dropout(0.5)(output)
    output = Dense(len(all_character_names), activation='softmax')(output)
    model = Model(model_base.input, output)
    for layer in model_base.layers:
        layer.trainable = False
    model.summary(line_length=200)

    # Generate a plot of a model
    import pydot
    pydot.find_graphviz = lambda: True
    from keras.utils import plot_model
    plot_model(model, show_shapes=True, to_file='../model_pdfs/{}.pdf'.format(pretrained_model))

    model.compile(optimizer='adam',
                  loss='categorical_crossentropy',
                  metrics=['accuracy'])
    return model