Sophisticate Open Confusion Matrix
Project description
conf-mat
This library was created to address the confusion in confusion matrices and classification reports generated by libraries like scikit-learn. It displays confusion matrices and their heatmaps in a more aesthetically pleasing, intuitive way. It also presents classification reports and accuracy rates in a manner that clarifies the calculation method and how those results were obtained, which can help alleviate confusion for beginners in machine learning
Installation
You can install conf-mat via pip:
pip install conf-mat
Usage
For Binary Classification
from numpy.random import randint
from matplotlib.pyplot import show
from conf_mat import print_conf_mat, plot_conf_mat, conf_mat_to_html
y_true = randint(0, 2, 1000)
y_pred = randint(0, 2, 1000)
# Show The Confusion Matrix On The Console (Classes Names Are Not Mandatory)
print_conf_mat(y_true, y_pred, classes_names=[False, True])
# Show The Confusion Matrix As A Heat Map (Classes Names Are Not Mandatory)
plot_conf_mat(y_true, y_pred, classes_names=[False, True])
show() # If You Are Using VSCode And You Are Executing The Code Direcctely
# Show The Confusion Matrix As It Appears On The Console And As A Heat Map On The HTML Page (Classes Names Are Not Mandatory)
conf_mat_to_html(y_true, y_pred, classes_names=[False, True])
Output
Confusion Matrix :
________________
╒═════════════════════╤═══════════════════════════╤═══════════════════════════╕
│ Classes │ Predicted Positive (PP) │ Predicted Negative (PN) │
╞═════════════════════╪═══════════════════════════╪═══════════════════════════╡
│ Actual Positive (P) │ True Positive (TP) : 273 │ False Negative (FN) : 244 │
│ │ │ Type II Error (Missed) │
├─────────────────────┼───────────────────────────┼───────────────────────────┤
│ Actual Negative (N) │ False Positive (FP) : 229 │ True Negative (TN) : 254 │
│ │ Type I Error (Wrong) │ │
╘═════════════════════╧═══════════════════════════╧═══════════════════════════╛
╒══════════╤══════════════════════════════════════════════════════╕
│ │ Rate (Score) │
╞══════════╪══════════════════════════════════════════════════════╡
│ Accuracy │ Correct TP + TN │
│ │ _______ : _________________ OR 1 - Error = 0.53 │
│ │ │
│ │ Total TP + FP + FN + TN │
├──────────┼──────────────────────────────────────────────────────┤
│ Error │ Wrong FP + FN │
│ │ _____ : _________________ OR 1 - Accuracy = 0.47 │
│ │ │
│ │ Total TP + FP + FN + TN │
╘══════════╧══════════════════════════════════════════════════════╛
Classification Report :
_____________________
╒══════════════════╤═════════════════╤════════════════════╤═════════════════════╤════════════════╕
│ │ Precision (P) │ Recall (R) │ F1-Score (F) │ Support (S) │
╞══════════════════╪═════════════════╪════════════════════╪═════════════════════╪════════════════╡
│ Positive (True) │ P1 (PPV): │ R1 (Sensitivity): │ F1 : │ S1 : │
│ │ │ │ │ │
│ │ TP │ TP │ 2 x P1 x R1 │ │
│ │ _______ = 0.54 │ _______ = 0.53 │ ___________ = 0.54 │ TP + FN = 517 │
│ │ │ │ │ │
│ │ TP + FP │ TP + FN │ P1 + R1 │ │
├──────────────────┼─────────────────┼────────────────────┼─────────────────────┼────────────────┤
│ Negative (False) │ P0 (NPV): │ R0 (Specificity): │ F0 : │ S0 : │
│ │ │ │ │ │
│ │ TN │ TN │ 2 x P0 x R0 │ │
│ │ _______ = 0.51 │ _______ = 0.53 │ ___________ = 0.52 │ FP + TN = 483 │
│ │ │ │ │ │
│ │ TN + FN │ TN + FP │ P0 + R0 │ │
├──────────────────┼─────────────────┼────────────────────┼─────────────────────┼────────────────┤
│ Macro Avg │ P1 + P0 │ R1 + R0 │ F1 + F0 │ TS = 1000 │
│ │ _______ = 0.53 │ _______ = 0.53 │ _______ = 0.53 │ │
│ │ │ │ │ │
│ │ 2 │ 2 │ 2 │ │
├──────────────────┼─────────────────┼────────────────────┼─────────────────────┼────────────────┤
│ Weighted Avg │ W1 │ W2 │ W3 │ TS = 1000 │
│ │ __ = 0.53 │ __ = 0.53 │ __ = 0.53 │ │
│ │ │ │ │ │
│ │ TS │ TS │ TS │ │
╘══════════════════╧═════════════════╧════════════════════╧═════════════════════╧════════════════╛
PPV : Positive Predictive Value
NPV : Negative Predictive Value
W1 = (P1 x S1) + (P0 x S0)
W2 = (R1 x S1) + (R0 x S0)
W3 = (F1 x S1) + (F0 x S0)
TS : Total Support = S1 + S0
Note : All Real Numbers Are Rounded With Two Digits After The Comma
You Can Plot, Generate HTML Page For Confusion Matrix And Classification Report Directly From One Calculation
from numpy.random import randint
from matplotlib.pyplot import show
from conf_mat import print_conf_mat, plot_conf_mat, conf_mat_to_html
y_true = randint(0, 2, 1000)
y_pred = randint(0, 2, 1000)
# Show The Confusion Matrix On The Console (Classes Names Are Not Mandatory)
cm = print_conf_mat(y_true, y_pred, classes_names=[False, True])
# Show The Confusion Matrix As A Heat Map (Classes Names Are Not Mandatory)
plot_conf_mat(conf_mat=cm, classes_names=[False, True])
show() # If You Are Using VSCode And You Are Executing The Code Direcctely
# Show The Confusion Matrix As It Appears On The Console And As A Heat Map On The HTML Page (Classes Names Are Not Mandatory)
conf_mat_to_html(conf_mat=cm, classes_names=[False, True])
You Can Hide Detail Information For Confusion Matrix, Classification Report And Confusion Matrix Heatmap
from numpy.random import randint
from matplotlib.pyplot import show
from conf_mat import print_conf_mat, plot_conf_mat, conf_mat_to_html
y_true = randint(0, 2, 1000)
y_pred = randint(0, 2, 1000)
# Show The Confusion Matrix On The Console (Classes Names Are Not Mandatory)
cm = print_conf_mat(y_true, y_pred, classes_names=[False, True], detail=False)
# Show The Confusion Matrix As A Heat Map (Classes Names Are Not Mandatory)
plot_conf_mat(conf_mat=cm, classes_names=[False, True], detail=False)
show() # If You Are Using VSCode And You Are Executing The Code Direcctely
# Show The Confusion Matrix As It Appears On The Console And As A Heat Map On The HTML Page (Classes Names Are Not Mandatory)
conf_mat_to_html(conf_mat=cm, classes_names=[False, True], detail=False)
Output
Confusion Matrix :
________________
╒═════════════════════╤═══════════════════════════╤═══════════════════════════╕
│ Classes │ Predicted Positive (PP) │ Predicted Negative (PN) │
╞═════════════════════╪═══════════════════════════╪═══════════════════════════╡
│ Actual Positive (P) │ 261 │ 271 │
├─────────────────────┼───────────────────────────┼───────────────────────────┤
│ Actual Negative (N) │ 240 │ 228 │
╘═════════════════════╧═══════════════════════════╧═══════════════════════════╛
╒══════════╤════════════════╕
│ │ Rate (Score) │
╞══════════╪════════════════╡
│ Accuracy │ 0.49 │
├──────────┼────────────────┤
│ Error │ 0.51 │
╘══════════╧════════════════╛
Classification Report :
_____________________
╒══════════════════╤═════════════════╤══════════════╤════════════════╤═══════════════╕
│ │ Precision (P) │ Recall (R) │ F1-Score (F) │ Support (S) │
╞══════════════════╪═════════════════╪══════════════╪════════════════╪═══════════════╡
│ Positive (True) │ 0.52 │ 0.49 │ 0.51 │ 532 │
├──────────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ Negative (False) │ 0.46 │ 0.49 │ 0.47 │ 468 │
├──────────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ Macro Avg │ 0.49 │ 0.49 │ 0.49 │ 1000 │
├──────────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ Weighted Avg │ 0.49 │ 0.49 │ 0.49 │ 1000 │
╘══════════════════╧═════════════════╧══════════════╧════════════════╧═══════════════╛
You Can Plot And Generate HTML Page For Confusion Matrix Directly From One Calculation And Without Print Confusion Matrix And Classification Report
from numpy.random import randint
from matplotlib.pyplot import show
from conf_mat import calc_conf_mat, plot_conf_mat, conf_mat_to_html
y_true = randint(0, 2, 1000)
y_pred = randint(0, 2, 1000)
# Show The Confusion Matrix On The Console (Classes Names Are Not Mandatory)
cm = calc_conf_mat(y_true, y_pred)
print(cm)
# Show The Confusion Matrix As A Heat Map (Classes Names Are Not Mandatory)
plot_conf_mat(conf_mat=cm, classes_names=[False, True], detail=False)
show() # If You Are Using VSCode And You Are Executing The Code Direcctely
# Show The Confusion Matrix As It Appears On The Console And As A Heat Map On The HTML Page (Classes Names Are Not Mandcatory)
conf_mat_to_html(conf_mat=cm, classes_names=[False, True], detail=False)
Output
[[236, 253], [255, 256]]
For Multi Classification
from numpy.random import randint
from matplotlib.pyplot import show
from conf_mat import print_conf_mat, plot_conf_mat, conf_mat_to_html
y_true = randint(0, 10, 1000)
y_pred = randint(0, 10, 1000)
# Show The Confusion Matrix On The Console (Classes Names Are Not Mandatory)
print_conf_mat(y_true, y_pred, classes_names=[
'C1', 'C2', 'C3', 'C4', 'C5', 'C6', 'C7', 'C8', 'C9', 'C10'])
# Show The Confusion Matrix As A Heat Map (Classes Names Are Not Mandatory)
plot_conf_mat(y_true, y_pred, classes_names=[
'C1', 'C2', 'C3', 'C4', 'C5', 'C6', 'C7', 'C8', 'C9', 'C10'])
show() # If You Are Using VSCode And You Are Executing The Code Direcctely
# Show The Confusion Matrix As It Appears On The Console And As A Heat Map On The HTML Page (Classes Names Are Not Mandatory)
conf_mat_to_html(y_true, y_pred, classes_names=[
'C1', 'C2', 'C3', 'C4', 'C5', 'C6', 'C7', 'C8', 'C9', 'C10'])
Output
Confusion Matrix :
________________
╒═══════════╤══════╤══════╤══════╤══════╤══════╤══════╤══════╤══════╤══════╤═══════╕
│ Classes │ C1 │ C2 │ C3 │ C4 │ C5 │ C6 │ C7 │ C8 │ C9 │ C10 │
╞═══════════╪══════╪══════╪══════╪══════╪══════╪══════╪══════╪══════╪══════╪═══════╡
│ C1 │ 5 │ 17 │ 12 │ 11 │ 12 │ 13 │ 6 │ 13 │ 10 │ 7 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C2 │ 10 │ 6 │ 10 │ 11 │ 13 │ 11 │ 13 │ 8 │ 11 │ 11 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C3 │ 6 │ 6 │ 12 │ 8 │ 10 │ 5 │ 11 │ 8 │ 7 │ 11 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C4 │ 14 │ 13 │ 10 │ 8 │ 9 │ 5 │ 12 │ 14 │ 9 │ 7 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C5 │ 12 │ 9 │ 15 │ 5 │ 8 │ 9 │ 6 │ 8 │ 10 │ 10 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C6 │ 12 │ 16 │ 15 │ 9 │ 10 │ 8 │ 10 │ 11 │ 11 │ 12 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C7 │ 12 │ 9 │ 12 │ 14 │ 12 │ 6 │ 13 │ 13 │ 7 │ 14 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C8 │ 6 │ 11 │ 7 │ 10 │ 10 │ 12 │ 8 │ 10 │ 8 │ 11 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C9 │ 15 │ 7 │ 8 │ 6 │ 8 │ 6 │ 5 │ 4 │ 14 │ 12 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C10 │ 10 │ 11 │ 11 │ 13 │ 8 │ 11 │ 12 │ 12 │ 6 │ 15 │
╘═══════════╧══════╧══════╧══════╧══════╧══════╧══════╧══════╧══════╧══════╧═══════╛
Yellow : Not None Correct Values / True Positive (TP) OR True Negative (TN)
Red : Not None Wrong Values / False Positive (FP) OR False Negative (FN)
Green : None Correct Values
Blue : None Wrong Values
╒══════════╤════════════════════════════════════════════════════════════════╕
│ │ Rate (Score) │
╞══════════╪════════════════════════════════════════════════════════════════╡
│ Accuracy │ Correct Sum Of Yellow Values │
│ │ _______ : ____________________________ OR 1 - Error = 0.1 │
│ │ │
│ │ Total Sum Of Yellow And Red Values │
├──────────┼────────────────────────────────────────────────────────────────┤
│ Error │ Wrong Sum Of Red Values │
│ │ _____ : ____________________________ OR 1 - Accuracy = 0.9 │
│ │ │
│ │ Total Sum Of Yellow And Red Values │
╘══════════╧════════════════════════════════════════════════════════════════╛
Classification Report :
_____________________
╒══════════════╤═════════════════╤══════════════╤════════════════╤═══════════════╕
│ │ Precision (P) │ Recall (R) │ F1-Score (F) │ Support (S) │
╞══════════════╪═════════════════╪══════════════╪════════════════╪═══════════════╡
│ C1 │ 0.05 │ 0.05 │ 0.05 │ 106 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C2 │ 0.06 │ 0.06 │ 0.06 │ 104 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C3 │ 0.11 │ 0.14 │ 0.12 │ 84 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C4 │ 0.08 │ 0.08 │ 0.08 │ 101 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C5 │ 0.08 │ 0.09 │ 0.08 │ 92 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C6 │ 0.09 │ 0.07 │ 0.08 │ 114 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C7 │ 0.14 │ 0.12 │ 0.12 │ 112 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C8 │ 0.1 │ 0.11 │ 0.1 │ 93 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C9 │ 0.15 │ 0.16 │ 0.16 │ 85 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C10 │ 0.14 │ 0.14 │ 0.14 │ 109 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ Macro Avg │ 0.1 │ 0.1 │ 0.1 │ 1000 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ Weighted Avg │ 0.1 │ 0.1 │ 0.1 │ 1000 │
╘══════════════╧═════════════════╧══════════════╧════════════════╧═══════════════╛
Precision : Yellow Value / Sum Of Yellow Value Column
Recall : Yellow Value / Sum Of Yellow Value Row
F1-Score : (2 x Precision x Recall) / (Precision + Recall)
Support : Sum Of Each Row
Macro Avg :
Precision : (Sum Of Precision Column) / Classes Count
Recall : (Sum Of Recall Column) / Classes Count
F1-Score : (Sum Of F1-Score Column) / Classes Count
Support : Total (Sum Of All Matrix)
Weighted Avg :
Precision : (Sum Of (Precision x support)) / Total (Sum Of All Matrix)
Recall : (Sum Of (Recall x Support)) / Total (Sum Of All Matrix)
F1-Score : (Sum Of (F1-Score x Support)) / Total (Sum Of All Matrix)
Support : Total (Sum Of All Matrix)
Note : All Real Numbers Are Rounded With Two Digits After The Comma
You Can Plot, Generate HTML Page For Confusion Matrix And Classification Report Directly From One Calculation
from numpy.random import randint
from matplotlib.pyplot import show
from conf_mat import print_conf_mat, plot_conf_mat, conf_mat_to_html
y_true = randint(0, 10, 1000)
y_pred = randint(0, 10, 1000)
# Show The Confusion Matrix On The Console (Classes Names Are Not Mandatory)
cm = print_conf_mat(y_true, y_pred, classes_names=[
'C1', 'C2', 'C3', 'C4', 'C5', 'C6', 'C7', 'C8', 'C9', 'C10'])
# Show The Confusion Matrix As A Heat Map (Classes Names Are Not Mandatory)
plot_conf_mat(conf_mat=cm, classes_names=[
'C1', 'C2', 'C3', 'C4', 'C5', 'C6', 'C7', 'C8', 'C9', 'C10'])
show() # If You Are Using VSCode And You Are Executing The Code Direcctely
# Show The Confusion Matrix As It Appears On The Console And As A Heat Map On The HTML Page (Classes Names Are Not Mandatory)
conf_mat_to_html(conf_mat=cm, classes_names=[
'C1', 'C2', 'C3', 'C4', 'C5', 'C6', 'C7', 'C8', 'C9', 'C10'])
You Can Hide Detail Information For Confusion Matrix, Classification Report And Confusion Matrix Heatmap
from numpy.random import randint
from matplotlib.pyplot import show
from conf_mat import print_conf_mat, plot_conf_mat, conf_mat_to_html
y_true = randint(0, 10, 1000)
y_pred = randint(0, 10, 1000)
# Show The Confusion Matrix On The Console (Classes Names Are Not Mandatory)
cm = print_conf_mat(y_true, y_pred, classes_names=[
'C1', 'C2', 'C3', 'C4', 'C5', 'C6', 'C7', 'C8', 'C9', 'C10'], detail=False)
# Show The Confusion Matrix As A Heat Map (Classes Names Are Not Mandatory)
plot_conf_mat(conf_mat=cm, classes_names=[
'C1', 'C2', 'C3', 'C4', 'C5', 'C6', 'C7', 'C8', 'C9', 'C10'], detail=False)
show() # If You Are Using VSCode And You Are Executing The Code Direcctely
# Show The Confusion Matrix As It Appears On The Console And As A Heat Map On The HTML Page (Classes Names Are Not Mandatory)
conf_mat_to_html(conf_mat=cm, classes_names=[
'C1', 'C2', 'C3', 'C4', 'C5', 'C6', 'C7', 'C8', 'C9', 'C10'], detail=False)
Output
Confusion Matrix :
________________
╒═══════════╤══════╤══════╤══════╤══════╤══════╤══════╤══════╤══════╤══════╤═══════╕
│ Classes │ C1 │ C2 │ C3 │ C4 │ C5 │ C6 │ C7 │ C8 │ C9 │ C10 │
╞═══════════╪══════╪══════╪══════╪══════╪══════╪══════╪══════╪══════╪══════╪═══════╡
│ C1 │ 10 │ 5 │ 7 │ 7 │ 8 │ 10 │ 9 │ 10 │ 12 │ 8 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C2 │ 7 │ 14 │ 14 │ 9 │ 9 │ 13 │ 14 │ 11 │ 14 │ 7 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C3 │ 13 │ 12 │ 5 │ 9 │ 6 │ 8 │ 9 │ 10 │ 14 │ 8 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C4 │ 6 │ 11 │ 9 │ 18 │ 11 │ 7 │ 10 │ 11 │ 6 │ 7 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C5 │ 9 │ 5 │ 11 │ 9 │ 7 │ 8 │ 8 │ 20 │ 9 │ 10 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C6 │ 12 │ 5 │ 8 │ 11 │ 11 │ 12 │ 14 │ 14 │ 11 │ 7 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C7 │ 13 │ 8 │ 10 │ 7 │ 10 │ 12 │ 15 │ 10 │ 8 │ 8 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C8 │ 10 │ 5 │ 6 │ 14 │ 11 │ 9 │ 9 │ 10 │ 4 │ 13 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C9 │ 11 │ 13 │ 10 │ 10 │ 13 │ 10 │ 8 │ 12 │ 9 │ 15 │
├───────────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼──────┼───────┤
│ C10 │ 11 │ 12 │ 11 │ 10 │ 9 │ 8 │ 6 │ 14 │ 12 │ 15 │
╘═══════════╧══════╧══════╧══════╧══════╧══════╧══════╧══════╧══════╧══════╧═══════╛
╒══════════╤════════════════╕
│ │ Rate (Score) │
╞══════════╪════════════════╡
│ Accuracy │ 0.12 │
├──────────┼────────────────┤
│ Error │ 0.89 │
╘══════════╧════════════════╛
Classification Report :
_____________________
╒══════════════╤═════════════════╤══════════════╤════════════════╤═══════════════╕
│ │ Precision (P) │ Recall (R) │ F1-Score (F) │ Support (S) │
╞══════════════╪═════════════════╪══════════════╪════════════════╪═══════════════╡
│ C1 │ 0.1 │ 0.12 │ 0.11 │ 86 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C2 │ 0.16 │ 0.12 │ 0.14 │ 112 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C3 │ 0.05 │ 0.05 │ 0.05 │ 94 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C4 │ 0.17 │ 0.19 │ 0.18 │ 96 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C5 │ 0.07 │ 0.07 │ 0.07 │ 96 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C6 │ 0.12 │ 0.11 │ 0.12 │ 105 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C7 │ 0.15 │ 0.15 │ 0.15 │ 101 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C8 │ 0.08 │ 0.11 │ 0.09 │ 91 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C9 │ 0.09 │ 0.08 │ 0.09 │ 111 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ C10 │ 0.15 │ 0.14 │ 0.15 │ 108 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ Macro Avg │ 0.12 │ 0.11 │ 0.11 │ 1000 │
├──────────────┼─────────────────┼──────────────┼────────────────┼───────────────┤
│ Weighted Avg │ 0.12 │ 0.12 │ 0.12 │ 1000 │
╘══════════════╧═════════════════╧══════════════╧════════════════╧═══════════════╛
You Can Plot And Generate HTML Page For Confusion Matrix Directly From One Calculation And Without Print Confusion Matrix And Classification Report
from numpy.random import randint
from matplotlib.pyplot import show
from conf_mat import calc_conf_mat, plot_conf_mat, conf_mat_to_html
y_true = randint(0, 10, 1000)
y_pred = randint(0, 10, 1000)
# Show The Confusion Matrix On The Console (Classes Names Are Not Mandatory)
cm = calc_conf_mat(y_true, y_pred)
print(cm)
# Show The Confusion Matrix As A Heat Map (Classes Names Are Not Mandatory)
plot_conf_mat(conf_mat=cm, classes_names=[
'C1', 'C2', 'C3', 'C4', 'C5', 'C6', 'C7', 'C8', 'C9', 'C10'], detail=False)
show() # If You Are Using VSCode And You Are Executing The Code Direcctely
# Show The Confusion Matrix As It Appears On The Console And As A Heat Map On The HTML Page (Classes Names Are Not Mandatory)
conf_mat_to_html(conf_mat=cm, classes_names=[
'C1', 'C2', 'C3', 'C4', 'C5', 'C6', 'C7', 'C8', 'C9', 'C10'], detail=False)
Output
[[9, 12, 11, 8, 10, 10, 7, 6, 11, 9], [12, 9, 12, 9, 7, 3, 16, 7, 7, 11], [7, 10, 25, 14, 5, 14, 9, 11, 6, 12], [7, 15, 9, 11, 13, 13, 4, 7, 10, 10], [12, 10, 11, 3, 14, 6, 8, 19, 6, 15], [5, 14, 13, 17, 9, 8, 8, 7, 11, 17], [7, 10, 10, 10, 6, 9, 10, 4, 9, 11], [8, 9, 11, 7, 11, 11, 15, 9, 13, 12], [12, 12, 9, 6, 11, 7, 10, 13, 7, 10], [11, 5, 9, 13, 8, 14, 3, 9, 15, 13]]
Note
These outputs are for the calc_conf_mat and print_conf_mat functions only, in both cases. I'll leave it to you to figure out the last two functions plot_conf_mat and conf_mat_to_html on your own.
You can call all the functions with the camelCase format.
License
This project is licensed under the MIT LICENSE - see the LICENSE for more details.
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
conf-mat-1.0.4.tar.gz
(21.2 kB
view hashes)
Built Distribution
conf_mat-1.0.4-py3-none-any.whl
(15.0 kB
view hashes)
