hjhhjhj

You plan to create a speech recognition deep learning model for your company.
The model must support the latest version of Python.
You need to recommend a deep learning framework for speech recognition to include in the Data Science Virtual Machine (DSVM).

What should you recommend?

• TensorFlow

• Rattle

• • Weka

• Scikit-learn

Answer explanation

Correct

TensorFlow is an open source library for numerical computation and large-scale machine learning. It uses Python to provide a convenient front-end API for building applications with the framework
TensorFlow can train and run deep neural networks for handwritten digit classification, image recognition, word embeddings, recurrent neural networks, sequence-to-sequence models for machine translation, natural language processing, and PDE (partial differential equation) based simulations.
Incorrect Answers:
§ Rattle is the R analytical tool that gets you started with data analytics and machine learning.
§ Weka is used for visual data mining and machine learning software in Java.
§ Scikit-learn is one of the most useful library for machine learning in Python. It is on NumPy, SciPy and matplotlib, this library
    contains a lot of efficient tools for machine learning and statistical modelling including classification, regression, and clustering and
    dimensionality reduction.

2. Question

You are a data scientist of your company and are asked implementing a machine learning model to predict stock prices.
The model uses a PostgreSQL database and requires GPU processing.
You need to create a virtual machine that is pre-configured with the required tools.

What should you do?

• Create a Geo Al Data Science Virtual Machine (Geo-DSVM) Windows edition.

• Create a Deep Learning Virtual Machine (DLVM) Windows edition.

• Create a Data Science Virtual Machine (DSVM) Windows edition.

• Create a Deep Learning Virtual Machine (DLVM) Linux edition.

Answer explanation

The best option for this scenario is a Deep Learning Virtual Machine (DLVM).

Here’s why:

• GPU Processing: DLVMs are specifically designed for workloads requiring GPUs, which are essential for deep learning models like stock price prediction.

• PostgreSQL Database: Both DLVMs and Data Science Virtual Machines (DSVMs) can likely handle PostgreSQL databases.

While Windows editions might be familiar to some users, Linux offers a wider range of data science tools and libraries, making DLM with Linux (option D) the most suitable choice.

So the answer is: D. Create a Deep Learning Virtual Machine (DLVM) Linux edition.

You are creating an experiment by using Azure Machine Learning Studio for your company.
Your task is to divide the data into four subsets for evaluation.
There is a high degree of missing values in the data. You must prepare the data for analysis.
You need to select appropriate methods for producing the experiment.
Which three modules should you run in sequence?

• Import Data Missing Values Scrubber Partition and Sample

• • Build Counting Transform Feature Hashing Partition and Sample

• Import Data Feature Hashing Replace Discrete Values

• Import Data Clean Missing Data Partition and Sample

You are a data scientist of a startup company.
You create an experiment in Azure Machine Learning Studio.
You add a training dataset that contains 10,000 rows. The first 9,000 rows represent class 0 (90 percent).
The remaining 1,000 rows represent class 1 (10 percent).
The training set is imbalances between two classes.
You must increase the number of training examples for class 1 to 4,000 by using 5 data rows.
You add the Synthetic Minority Oversampling Technique (SMOTE) module to the experiment.
You need to configure the module.
Which values should you use?

• SMOTE percentage: 4000 Number of nearest neighbours: 4000

• SMOTE percentage: 300 Number of nearest neighbours: 5

• SMOTE percentage: 300 Number of nearest neighbours: 1

• SMOTE percentage: 3000 Number of nearest neighbours: 5

Answer explanation

Nr vecini = in enunt

cu cat % creste procentul = smote percentage.

in cazul asta vrem sa avem 4000 de la 1000 => nr initial creste cu 300%

Studiu de caz sporting event.

You need to use the Python language to build a sampling strategy for the global penalty detection models.
How should you complete the code segment?

• Box1: import pytorch as deeplearninglib Box2: train_sampler=deeplearninglib.DistributedSampler.(penalty_video_dataset) Box3: optimizer= deeplearninglib.optim.SGD(model.parameters().lr=0,01) Box4: model= deeplearninglib.keras.Model([

• Box1: import tensorflow as deeplearninglib Box2: train_sampler=deeplearninglib.log_uniform_candidate_sampler.(penalty_video_dataset) Box3: optimizer= deeplearninglib.optim.SGD(model.parameters().lr=0,01) Box4: model= deeplearninglib.nn.parallel.DistributedDataParallelCPU(model)

• Box1: import pytorch as deeplearninglib Box2: train_sampler=deeplearninglib.DistributedSampler.(penalty_video_dataset) Box3: optimizer= deeplearninglib.train.GradientDescentOptimizer(learning_rate=0.01) Box4: model= deeplearninglib.nn.parallel.DistributedDataParallelCPU(model)

• Box1: import cntk as deeplearninglib Box2: train_sampler=deeplearninglib.WeightedRandomSampler.(penalty_video_dataset) Box3: optimizer= deeplearninglib.train.GradientDescentOptimizer(learning_rate=0.01) Box4: model= deeplearninglib.parallel.Distributed(DataParallel(model)

Answer explanation

• Box 1 (import pytorch as deeplearninglib): This choice is correct because PyTorch is a popular deep learning library that supports dynamic computation graphs, which are required for global penalty detection models.

• Box 2 (train_sampler = deeplearninglib.DistributedSampler(penalty_video_dataset)): This line uses a DistributedSampler, which is ideal for splitting data across multiple devices (e.g., GPUs) in a distributed environment. Since the problem involves handling large-scale media data from penalty detection models, distributing the data across devices ensures faster training.

• Box 3 (optimizer = deeplearninglib.optim.SGD(model.parameters(), lr=0.01)): The SGD (Stochastic Gradient Descent) optimizer is often used in distributed settings, especially when performance is critical. The learning rate of 0.01 matches the requirements to handle high-bias, low-variance models efficiently.

• Box 4 (model = deeplearninglib.nn.parallel.DistributedDataParallelCPU(model)): This is the correct choice because DistributedDataParallel is a PyTorch utility for parallelizing models across multiple devices or CPUs. Since this project requires training global penalty detection models in a distributed environment, this function ensures that the model's computations are distributed across the available CPU resources efficiently.

You need to normalize values to produce an output column into bins to predict a target column using Azure Machine Learning Studio.
Solution:
Apply an Equal Width with Custom Start and Stop binning mode.

Does the solution meet the goal?

• Yes

• No

Answer explanation

You need to use the Entropy MDL binning mode which has a target column.

You are a data scientist in a company that provides data science for professional sporting events. Models will use global and local market data to meet the following business goals:You need to implement a model development strategy to determine a user’s tendency to respond to an ad.
Which technique should you use?

• Use a Relative Expression Split module to partition the data based on distance travelled to the event

• Use a Relative Expression Split module to partition the data based on centroid distance

• Use a Split Rows module to partition the data based on distance travelled to the event

• Use a Split Rows module to partition the data based on centroid distance

Your manager asked you to use a Data Science Virtual Machine (DSVM) with the open source deep learning frameworks Caffe2 and PyTorch.
You need to select a pre-configured DSVM to support the frameworks.
What should you create?

• • Data Science Virtual Machine for Windows 2012

• • Data Science Virtual Machine for Linux (Ubuntu)

• • Data Science Virtual Machine for Windows 2016

• • Geo AI Data Science Virtual Machine with ArcGIS

• • Option D

• Option A

• Option C

• Option E

You define a datastore named ml-data for an Azure Storage blob container. In the container, you have a folder named train that contains a file named data.csv. You plan to use the file to train a model by using the Azure Machine Learning SDK.
You plan to train the model by using the Azure Machine Learning SDK to run an experiment on local compute.
You define a DataReference object by running the following code:

• 
• 
• 
• 
•