The IIT Madras BS Machine Learning Practice (MLP) Quiz 1 paper sat on 26 Feb 2023, in the January 2023 term: 23 questions for 50 marks in 120 minutes. Every question is below with its answer. Take it as a timed mock test to be marked, or read it through first.
[String Matching] Enter the sequence of steps to be followed, in general, in end to end machine learning project. Enter the answer as a 6-character string. For example, entering the answer as (without quotes) ‘BADCFE’ implies that the first step is B, followed by the second step A and so on. A. Select suitable model
B. Train the model
C. Pre-process the data
D. Collect data
E. Fine tune model
F. Present your solution
Correct answer: DCABEF
Which of the following APIs can be used to construct an ML pipeline for data preprocessing and modeling?
Pipeline
ColumnTransformer
FeatureUnion
All of these
Correct answer
All of these
Consider following data:
Which one of the following APIs can be used to extract features from the above data?
DictVectorizer
HashingVectorizer
FeatureHasher
Correct answer
DictVectorizer
Consider following data:
What will be the output of the following code? (Assume necessary imports)
si = SimpleImputer(strategy='mean')Xnew = si.fit_transform(X)print(f'{Xnew[1,0]}, {Xnew[2,1]}')7,6
6,6
6,7
7,7
Correct answer
7,6
Suppose that we load a data set that contains 1000 samples in a Pandas Dataframe. Each sample has 30 features. However, a few samples in the data set miss the values for all features. Therefore, those samples need to be dropped. Choose the method that removes such samples from the dataset?
drop(columns=[‘all’])
drop(how=‘all’)
dropna()
dropna(how=‘all’)
dropna.all()
Correct answer
dropna(how=‘all’)
The statement that each feature in the input data set has a (physical) meaning associated with it is
True, for all ML problems
True, for some ML problems
False, for all ML problems
can not be decided
Correct answer
True, for some ML problems
A company collects 40000 samples (examples) to build a Machine Learning model for an application. They decide to use 30% of the total samples for testing (to be stored in the variable trainset) and the rest 70% for training (to be stored in the variable trainset). They also want to sample the same set of samples across multiple runs. Which of the following line (statement) achieves this task? Assume that all samples are stored in the variable data.
Correct answer
Which of the following utilities of sklearn.datasets helps us to get the realworld data from the internet?
load_
fetch_
generate_
get_
Correct answer
fetch_
Consider the following ML task/steps for a regression dataset:
1. Read the data from a file (named ‘dataset.csv’). It has 7 columns. The last column is the target variable, the rest of them are numerical features.
2. Drop rows with missing values.
3. After removing samples with missing values split the data into training and test sets. Take about the first 80% of rows in the training set and the rest of them into the test set.
4. Train a simple linear regression model, with intercept, on the training set.
5. Report R2 score on the test set.
Which of the following code snippets correctly accomplishes the above task? Assume necessary imports.
Correct answer
Consider the following code where X_train, y_train is the training data. X_test, y_test is the test data.
Which evaluation metric will be contained in the ‘score’?
mean_squared_error
mean_absolute_error
R2_score
Accuracy
Correct answer
R2_score
How to make SGDRegressor stop after 1000 epochs?
Correct answer
Suppose we want to transform features in a dataset using polynomial transformation. The sklearn API provides the functionality in which of the following modules?
sklearn.dataset
sklearn.model_selection
sklearn.preprocessing
sklearn.featureSelection
sklearn.featureExtraction
Correct answer
sklearn.preprocessing
Consider following dataset:
Which one of the following code snippets will correctly preprocess above data? Assume necessary imports. The data is stored in a dataframe named X.
Correct answer
For a dataset with 1000 data points and 100 features, the following code will generate how many models during execution?
Note: X is the feature matrix and y is the target vector.
1000
100
99
999
Correct answer
1000
Consider the following cross-validation strategy:
Assume we apply this strategy to some data set. Which of the following options is/are correct?
Correct answer
Which of the following code is correct if we want the mean absolute error to be minimized during the computation of cross_val_score?
Correct answer
We know that applying polynomial transformation to the samples is often helpful. Assume we imported all the required modules.
(2,2)
(2,6)
(6,2)
(2,3)
(3,2)
(2,5)
(5,2)
Correct answer
(2,3)
A team has built a dataset for a regression problem. It contains 1000 samples. Each sample x is of size 2. All the features are binary, that is, xi ∈ {0, 1}. The team decided to use polynomial feature transformation of degree 2 as follows,
1
2
3
4
0
Correct answer
2
Consider a regression problem with L2 regularization. Suppose we instantiate the model as shown below,
What is the range of ‘alpha’?
Correct answer
(Multiple Select) Consider a data set shown below. The data set was loaded using Pandas and stored in the variable ‘data’ as a data frame.
| fixed acidity | volatile acidity | citric acid | residual sugar | chlorides | free sulfur dioxide | total sulfur dioxide | density | pH | sulphates | alcohol | quality | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 7.4 | 0.70 | 0.00 | 1.9 | 0.076 | 11.0 | 34.0 | 0.9978 | 3.51 | 0.56 | 9.4 | 5 |
| 1 | 7.8 | 0.88 | 0.00 | 2.6 | 0.098 | 25.0 | 67.0 | 0.9968 | 3.20 | 0.68 | 9.8 | 5 |
| 2 | 7.8 | 0.76 | 0.04 | 2.3 | 0.092 | 15.0 | 54.0 | 0.9970 | 3.26 | 0.65 | 9.8 | 5 |
| 3 | 11.2 | 0.28 | 0.56 | 1.9 | 0.075 | 17.0 | 60.0 | 0.9980 | 3.16 | 0.58 | 9.8 | 6 |
| 4 | 7.4 | 0.70 | 0.00 | 1.9 | 0.076 | 11.0 | 34.0 | 0.9978 | 3.51 | 0.56 | 9.4 | 5 |
data_url = '''https://archive.ics.uci.edu/ml/machine-learning-databases/wine-quality/winequality-red.csv'''data = pd.read_csv(data_url, sep=";")Suppose that we want to get the value of ‘chlorides‘ of the third sample (2nd by index). Which of the following lines of code does this?
data.chlorides[2]
data[‘chlorides’][2]
data[2][4]
data.iloc[2,4]
Correct answers
data.chlorides[2]
data[‘chlorides’][2]
data.iloc[2,4]
(MSQ) Suppose that we plot the histogram of numerical features in a data set. This reveals which of the following information?
Scale of the features
(left or right) Skew of the distribution
Modes in the distribution
Deduce total number of samples in the dataset
Correct answers
Scale of the features
(left or right) Skew of the distribution
Modes in the distribution
Deduce total number of samples in the dataset
Why is data preprocessing necessary?
While recording or noting down, the data collector forgot the values to be recorded and entered blanks.
Some columns have values only between 0 and 1.
The data is divided into multiple files and has to be combined.
The data has only numbers in all the columns.
Correct answers
While recording or noting down, the data collector forgot the values to be recorded and entered blanks.
The data is divided into multiple files and has to be combined.
[MSQ] Which of the following will produce constantly reducing learning rates?
Correct answers