ML Interview at Université Ibn Zohr Agadir | Flashcards & Summaries

Select your language

Suggested languages for you:
Log In Start studying!

Lernmaterialien für ML interview an der Université Ibn Zohr Agadir

Greife auf kostenlose Karteikarten, Zusammenfassungen, Übungsaufgaben und Altklausuren für deinen ML interview Kurs an der Université Ibn Zohr Agadir zu.

TESTE DEIN WISSEN

Feature Extraction 


Lösung anzeigen
TESTE DEIN WISSEN

A slightly less common, more specialized approach to deep learning is to use the network as a feature extractor. Since all the layers are tasked with learning certain features from images, we can pull these features out of the network at any time during the training process. These features can then be used as input to a machine learning model such as support vector machines (SVM).

Lösung ausblenden
TESTE DEIN WISSEN

3. Fit the model

Lösung anzeigen
TESTE DEIN WISSEN
  • the fit function(code) training the model for a fixed number of epochs (iterations on a datasets)
  • important parameters : 
    • number of epochs
    • input and output data
    • validation data 
  • the fitting step must be continuously evaluated while the training procedure is underway
  • make sure the model trained well with improving accuracies and reduction in overall loss
  • Tensorboard must be used for analyzing the various graphs and understanding if these models are being overfitted by any change
Lösung ausblenden
TESTE DEIN WISSEN
Stock market ?
Lösung anzeigen
TESTE DEIN WISSEN
Stock market is a place where buying and selling and selling of shares happen to publicly listed companies
Lösung ausblenden
TESTE DEIN WISSEN

What is an API?

Lösung anzeigen
TESTE DEIN WISSEN

An application programming interface is a connection between computers or between computer programs.

Lösung ausblenden
TESTE DEIN WISSEN

​4. Regularization

Lösung anzeigen
TESTE DEIN WISSEN

Regularization reduce overfitting by adding penalty to all features which are useless


Dropout is an approach of regularization in NN

  • Robust features
  • reduce training time
  • chose the right value gives good results

value of dropout vary from 0 to 1

0 => no dropout

0.2 => 20% of all neurons from hidden layer is removed

Lösung ausblenden
TESTE DEIN WISSEN

1. Activation function 

Lösung anzeigen
TESTE DEIN WISSEN

transform the data provided by input layer into output layer that are needed to a NN to function


  • Linear actv function :

f(x) = x ; do not help with complexity of problem

  • non L actv function : 

most used makes it easy for the model to adapt or generalize with a varity of data + derivative

Lösung ausblenden
TESTE DEIN WISSEN
Stock merket prediction
Lösung anzeigen
TESTE DEIN WISSEN
Stock market prediction helps you determine the future value of company stock and other financial instruments traded on an extchange
Physical and psychological factors
Lösung ausblenden
TESTE DEIN WISSEN

2. Gradient descent 

Lösung anzeigen
TESTE DEIN WISSEN

Gradient descent is an optimization algorithm which is commonly-used to train machine learning models and neural networks.


GD algo:

  • ​bach
  • stochastix
  • mini batch
Lösung ausblenden
TESTE DEIN WISSEN

3. Optimizers ?

Lösung anzeigen
TESTE DEIN WISSEN

help to faster convergence and better rates.

​Ex: Momentum, RMS, Adam

they imporve the performance of training a specific model

Lösung ausblenden
TESTE DEIN WISSEN

4.1 Dropout?

Lösung anzeigen
TESTE DEIN WISSEN

droping or removing neurons at random hidden layer

Lösung ausblenden
TESTE DEIN WISSEN

when to use sigmoid fun and softmax?

Lösung anzeigen
TESTE DEIN WISSEN

sig for binary classification

softmax for multiple class classification

Lösung ausblenden
TESTE DEIN WISSEN
Importance of stock market 

Lösung anzeigen
TESTE DEIN WISSEN
  • Helps companies to raise capital
  • Help create personal wealth
  • Serves as an indicator of the state of the economy
  • Helps increase investment
Lösung ausblenden
  • 693 Karteikarten
  • 328 Studierende
  • 1 Lernmaterialien

Beispielhafte Karteikarten für deinen ML interview Kurs an der Université Ibn Zohr Agadir - von Kommilitonen auf StudySmarter erstellt!

Q:

Feature Extraction 


A:

A slightly less common, more specialized approach to deep learning is to use the network as a feature extractor. Since all the layers are tasked with learning certain features from images, we can pull these features out of the network at any time during the training process. These features can then be used as input to a machine learning model such as support vector machines (SVM).

Q:

3. Fit the model

A:
  • the fit function(code) training the model for a fixed number of epochs (iterations on a datasets)
  • important parameters : 
    • number of epochs
    • input and output data
    • validation data 
  • the fitting step must be continuously evaluated while the training procedure is underway
  • make sure the model trained well with improving accuracies and reduction in overall loss
  • Tensorboard must be used for analyzing the various graphs and understanding if these models are being overfitted by any change
Q:
Stock market ?
A:
Stock market is a place where buying and selling and selling of shares happen to publicly listed companies
Q:

What is an API?

A:

An application programming interface is a connection between computers or between computer programs.

Q:

​4. Regularization

A:

Regularization reduce overfitting by adding penalty to all features which are useless


Dropout is an approach of regularization in NN

  • Robust features
  • reduce training time
  • chose the right value gives good results

value of dropout vary from 0 to 1

0 => no dropout

0.2 => 20% of all neurons from hidden layer is removed

Mehr Karteikarten anzeigen
Q:

1. Activation function 

A:

transform the data provided by input layer into output layer that are needed to a NN to function


  • Linear actv function :

f(x) = x ; do not help with complexity of problem

  • non L actv function : 

most used makes it easy for the model to adapt or generalize with a varity of data + derivative

Q:
Stock merket prediction
A:
Stock market prediction helps you determine the future value of company stock and other financial instruments traded on an extchange
Physical and psychological factors
Q:

2. Gradient descent 

A:

Gradient descent is an optimization algorithm which is commonly-used to train machine learning models and neural networks.


GD algo:

  • ​bach
  • stochastix
  • mini batch
Q:

3. Optimizers ?

A:

help to faster convergence and better rates.

​Ex: Momentum, RMS, Adam

they imporve the performance of training a specific model

Q:

4.1 Dropout?

A:

droping or removing neurons at random hidden layer

Q:

when to use sigmoid fun and softmax?

A:

sig for binary classification

softmax for multiple class classification

Q:
Importance of stock market 

A:
  • Helps companies to raise capital
  • Help create personal wealth
  • Serves as an indicator of the state of the economy
  • Helps increase investment
ML interview

Erstelle und finde Lernmaterialien auf StudySmarter.

Greife kostenlos auf tausende geteilte Karteikarten, Zusammenfassungen, Altklausuren und mehr zu.

Jetzt loslegen

Das sind die beliebtesten ML interview Kurse im gesamten StudySmarter Universum

Internship interviews

Central Philippine University

Zum Kurs
CS questions for MBA interview

Bharati Vidyapeeth University

Zum Kurs
Interview Questions

North Carolina State University

Zum Kurs
inter

Universidad de Las Palmas de Gran Canaria

Zum Kurs

Die all-in-one Lernapp für Studierende

Greife auf Millionen geteilter Lernmaterialien der StudySmarter Community zu
Kostenlos anmelden ML interview
Erstelle Karteikarten und Zusammenfassungen mit den StudySmarter Tools
Kostenlos loslegen ML interview