Softmax Regression || Multinomial Logistic Regression || Logistic Regression Part 6

Softmax Regression, also known as Multinomial Logistic Regression, is an extension of logistic regression to handle multiple classes. It computes the probabilities of each class and selects the one with the highest probability as the predicted class. Softmax activation ensures that the predicted probabilities sum up to 1, providing a robust framework for multiclass classification tasks.

============================
Do you want to learn from me?
============================

📱 Grow with us:

⌚Time Stamps⌚

00:00 - Intro
00:55 - Softmax Regression
08:20 - Training Intuition
14:53 - Prediction
19:58 - Loss Function
34:30 - Softmax Demo in SKlearn

[raghavagrawal]

No one explains such things in deep and in an easy manner anywhere in a course or on youtube. Thank you so much sir for this level of explanation.

[varunahlawat]

I subscribed, this channel is
I've not seen a single other channel like this, its been more than 4 months of learning I just wish I would've found this earlier!

[tushargogiya]

Thank you so much sir for this level of explanation.

[varunahlawat]

This video was really hard to find, I was about to give up searching the mathematics behind multinomial logistic regression! Thankyou!

[rohitekka]

Your explanation are gold sir!
Thank you for video!

[Pipython]

Thank youuuu for this easy explanation . You are on top...

[katw]

thanks Sir please Complete Mathmatical Part also

[harsh.gupta]

But sir in previous video, you used LogisticRegression() class on mnist dataset which has 10 labels ( output ). But here u say that LogisticRegression is only able to perform Binary Classification Plzz

[prasadbhandarkar]

Sir please algorithm continue karo xgboost and gboost

[meetbhadeshia]

Hi why did we just use two columns for x, when we had 4? Please answer

[VishalKumar-choj]

Bhaiya
I run this but 😶😶😶😶
from matplotlib import axis
import matplotlib.pyplot as plt
from mlxtend.plotting import plot_decision_regions

plot_decision_regions(X, y, clf, legend=2)
# Adding axes annotations
plt.xlabel('sepal length [cm]')
plt.ylabel('petal length [cm]')
plt.title('Softmax on Iris')
plt.show()

getting error. Not able solve this.

TypeError: axis() got an unexpected keyword argument 'y_min'

[eeshananand]

Sir, I have a doubt since the dataset is not provided in the Github, I tried making classification model using the command:-
X, y = make_classification(n_samples=100, n_features=2, n_informative=1, n_redundant=0,
n_classes=3, n_clusters_per_class=1, random_state=41, hypercube=False, class_sep=20)

But seems like it gives me error, how can I deal with this?

[ShreyeshSharma]

Dont we use LabelEncoder for output col? as mentioned by you in some other video where you also showed the official scikit learn doc

[Philgob]

brudda please make whole video in english it’s hard to follow when you keep switching

[sans_moi]

can u please provide your onenote it will be easier for us to revise

[fullthrottlevishal]

nitish sir, species got three classess in it but why there is only 1 feature is there after ohe on it, in training module said after ohe if three classess were there so three column will be there kindly confirm that thing

[abhishekkukreja]

while doing logistic regression for each label, we're using sigmoid function right? I mean to ask whether while we train we don't use softmax ??

[oprotectordm]

31:40 formula will apply at one row ?? so X11+ X12+

[mohammadali-yyop]

Anyone can have implementation of logistic regression for multiclass

[tushargogiya]

have you post math intituion or scratch code if yes pls give me link