Scribd
Upload
11 views4 pages

Sheet - Evaluation Metrics and Introduction To Deep Learning

The document outlines the evaluation of a classification model, including the construction of a confusion matrix and calculations for precision, recall, accuracy, and F1 score. It also discusses regression metrics such as MAE, MSE, RMSE, and R-squared, and explains deep convolutional neural networks. Additionally, it includes true/false statements regarding deep learning concepts and practices.

Uploaded by

david1milad1982
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
11 views4 pages

Sheet - Evaluation Metrics and Introduction To Deep Learning

The document outlines the evaluation of a classification model, including the construction of a confusion matrix and calculations for precision, recall, accuracy, and F1 score. It also discusses regression metrics such as MAE, MSE, RMSE, and R-squared, and explains deep convolutional neural networks. Additionally, it includes true/false statements regarding deep learning concepts and practices.

Uploaded by

david1milad1982
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Sheet (Evaluation metrics and Deep learning)

1) The data in table describe the results of model A classification.

a) Create the confusion matrix and evaluate Model A


b) Fill in the values for these terms using the numbers above:

c) The accuracy of the classifier is??


d) What is the formula for recall in terms of TP, TN, FP, and FN?
e) What is formula for precision in terms of TP, TN, FP, and FN?
f) The numerical value of recall is and precision is??
g) If you always predict the majority class what will be the accuracy and
recall? Fill in below.

2) A binary classification model was tested on a dataset of 100 samples. The


results are as follows:
• Predicted Positive: 40
• True Positives (TP): 30
• False Positives (FP): 10
• Predicted Negative: 60
• True Negatives (TN): 50
• False Negatives (FN): 10
a) Construct the confusion matrix.
b) Calculate the following:
• Precision
• Recall
• Accuracy
• F1 Score
• Specificity
3) A regression model is tested on a dataset of 5 samples. The actual and
predicted values are as follows:

Calculate the following regression metrics:


• Mean Absolute Error (MAE)
• Mean Squared Error (MSE)
• Root Mean Squared Error (RMSE)
• R-squared (R^2)
4) Explain in detail the Deep Convolutional Neural Networks.

5) Consider a 4x4 input matrix and a 2x2 filter. Perform a convolution


operation with a stride of 1 and 0 padding. The input matrix and
filter are as follows:
Input Matrix:

Filter:

6) If a CNN has an input image of size 32×32, uses a 3×3 filter, a stride
of 2, and no padding, what will be the size of the resulting feature
map?
7) Let’s consider a simple Convolutional Neural Network (CNN)
operation where we apply a convolution followed by the ReLU
activation function. Below are the details.
8) True/False:
a) Deep learning models always outperform traditional machine
learning models.
b) ReLU is a non-linear activation function commonly used in deep
learning.
c) Overfitting in deep learning occurs when the model performs well
on training data but poorly on unseen data.
d) Dropout is a technique used to reduce overfitting in deep learning
models.
e) Deep learning models can only be trained using GPUs.
f) In supervised learning, deep learning models do not require
labeled data.
g) Accuracy is always the best metric to evaluate a classification
model.
h) Precision measures the proportion of correctly identified positive
predictions out of all positive predictions made.
i) Recall is also known as sensitivity or the true positive rate.
j) The F1-score is the harmonic mean of precision and recall.
k) The confusion matrix is only applicable to binary classification
problems.
l) Mean Squared Error (MSE) is used to evaluate classification
models.
m) A high precision score guarantees a high recall score.

You might also like