stud
bud
Sign up
Log in
Home
Library
Explore
SAT
More
Evaluation (free point) Flashcards | StudBud
Explore
/
Evaluation (free point)
What is fine-tuning?
What does fit_transform do?
Evaluation (free point)
25 terms
·
StudBud Administration
·
13 plays
Imported from external platform
Flashcards
25 cards
F
Practice quiz
P
Host live
L
Questions
25
Answers
1
What does
fit_transform
do?
It fits a model to the data and then transforms the data in one step
It applies scaling to your data
It trains a machine learning model and evaluates its performance
It only applies transformation to the data without fitting any model
2
What is fine-tuning?
Technique used to reduce the complexity of a model
Process of making adjustments to an existing model
A method of initializing the weights of a neural network from scratch
The process of scaling data before feeding it into a model
3
Which is not a metric for classification?
Precision and Recall
Hamming Loss
F1 score
Mean Squared Error
4
For what is attention important?
It prevents overfitting by limiting the complexity of the model
It speeds up the training process by reducing the size of the input data
It ensures that all input data points are processed equally
Helps models focus on the most relevant parts of the input sequence
5
What is attention?
Mechanism to focus on different parts of the input sequence
A process of normalizing data before feeding it into a neural network
Model architecture designed to handle sequential data
Initialization of the weights of a neural network from scratch
6
Which model is known for vanishing and exploding?
Support Vector Machines (SVMs)
K-Nearest Neighbors (KNN)
Deep Neural Networks (DNN)
Recurrent Neural Networks (RNNs)
7
What is a transformer?
A model used for clustering unstructured data
Model architecture to handle sequential data using self-attention mechanism
A method to optimize weights during backpropagation
A type of activation function in neural networks
8
Just the
.transform
, you apply on:
Test Set
Training Set
9
You can have a regressor model using NN
True
False
10
What is Entropy?
A metric used to evaluate linear regression models
The accuracy of a machine learning model
A measure of the disorder or randomness
Measure of computational power needed
11
Which is not a metric for regression?
RMSE
R-squared
F1 score
MAE
12
What is RAG?
A method to retrieve data from external sources to enhance generation tasks
Method for optimizing deep learning models using gradient-based approaches
A type of algorithm used to perform clustering in machine learning
A process of training models using reinforcement learning techniques
13
What characteristic does a model need to have to run on a GPU?
Compatibility with cloud computing platforms
Sequential operations
Operations that can be parallelized
A small number of parameters to reduce memory usage
14
How can you have a regressor model with NN?
Using SoftMax activation function
using a linear activation function
Using ReLU activation function
if you use a support vector machine (SVM) model with neural networks
15
What exactly do we need to take care with the last layer of a neural network?
Using dropout to prevent overfitting
Ensuring the layer has the same number of neurons as the input layer
Choosing the correct activation function based on the type of problem
Normalizing the output to ensure values are between 0 and 1
16
You should use
fit_transform
for training and test set
True
False
17
What is activation function?
A type of data preprocessing step used in machine learning
A process to evaluate model accuracy during training
A method for optimizing the weights of a model
A mathematical function that determines the output of a neural network node
18
Any model can benefit from the speed provided by GPUs
True
False
19
What is normalization?
The process of scaling input data to a specific range
A technique used to normalize a neural network
A technique for reducing the dimensionality of datasets
A process of optimizing the weights of a model
20
What convolution is?
Partial derivatives
Type of layer(s) in NN
A process to calculate the entropy of a dataset
Line detector in image datasets
21
Fit_transform
should be used
Just on Training Set
Just on Test Set
22
You can save your
fit_transform
model
False
True
23
What convolution does?
Detect lines on the image dataset
Minimize the loss function
Extract features from the image dataset
Control the input and output size between layers
24
Which model came to solve the vanishing/exploding problem?
Long Short-Term Memory (LSTM) networks
Convolutional Neural Networks (CNNs)
Generative Adversarial Networks (GANs)
Radial Basis Function (RBF) networks
25
What are embeddings?
Numerical representations of objects in a continuous vector space
A technique for reducing the dimensionality of datasets
A type of activation function used in neural networks
A method for optimizing the performance of a classification model
