import numpy as np
import pandas as pd
from sklearn.metrics import accuracy_score, recall_score, f1_score, precision_score


def training_overview(correct, predictions):
    print("Accuracy: ", accuracy_score(correct, predictions))
    # za ako se povekje od 2 klasi average='macro'.
    # labels=np.unique(predictions) dodadeno vo sluchaj da ima nekoi labeli vo correct koi gi nema vo predictions
    print("Precision: ", precision_score(correct, predictions, average='macro', labels=np.unique(predictions)))
    print("Recall score: ", recall_score(correct, predictions, average='macro'))
    print("F1 score: ", f1_score(correct, predictions, average='macro', labels=np.unique(predictions)))
    print()


def create_dict():
    classes_pd = pd.read_csv("./visual_genome_objects.csv", sep=',')
    classes = dict()

    for row in classes_pd.values:
        classes[row[0]] = row[1]

    return classes


def create_classes_one_hot_encoded(classes_dict):

    one_hot_encoded = dict()

    for key, value in classes_dict.items():
        if value == 'tree':
            one_hot_encoded[key] = [1, 0, 0, 0, 0, 0, 0, 0, 0, 0]
        elif value == 'car':
            one_hot_encoded[key] = [0, 1, 0, 0, 0, 0, 0, 0, 0, 0]
        elif value == 'man':
            one_hot_encoded[key] = [0, 0, 1, 0, 0, 0, 0, 0, 0, 0]
        elif value == 'road':
            one_hot_encoded[key] = [0, 0, 0, 1, 0, 0, 0, 0, 0, 0]
        elif value == 'building':
            one_hot_encoded[key] = [0, 0, 0, 0, 1, 0, 0, 0, 0, 0]
        elif value == 'window':
            one_hot_encoded[key] = [0, 0, 0, 0, 0, 1, 0, 0, 0, 0]
        elif value == 'desk':
            one_hot_encoded[key] = [0, 0, 0, 0, 0, 0, 1, 0, 0, 0]
        elif value == 'chair':
            one_hot_encoded[key] = [0, 0, 1, 0, 0, 0, 0, 1, 0, 0]
        elif value == 'shelf':
            one_hot_encoded[key] = [0, 0, 1, 0, 0, 0, 0, 0, 1, 0]
        elif value == 'glass':
            one_hot_encoded[key] = [0, 0, 1, 0, 0, 0, 0, 0, 0, 1]

    return one_hot_encoded