Purchasing model

 A SQL Server running in an Azure virtual machine (IaaS) is equivalent to an on-premises SQL Server. You'll notice that several features described for SQL Server on Azure virtual machine are applicable to all your on-premises SQL Servers.

Many applications require SQL Server running on a virtual machine. The reasons include:

• General application support and incompatibility - For applications requiring an older version of SQL Server for vendor support. In addition, some application services may have a requirement to be installed with the database instance in a manner that isn't compatible with a PaaS offering.

• Use of other SQL Server Services - In order to maximize licensing, many users choose to run SQL Server Analysis Services (SSAS), SQL Server Integration Services (SSIS), and/or SQL Server Reporting Services (SSRS) on the same machine as the database engine.

Versions of SQL Server available

Microsoft keeps images of all supported versions of SQL Server available in the Azure Marketplace. If you require an older version that is covered by an extended support contract, you'll need to install your own SQL Server binaries.

Backup solutions

In recent releases of SQL Server, Microsoft has introduced several features to support running SQL Server in an Azure virtual machine. We're going to focus on two key backup features:

• Back up to URL
• Azure Backup

The back up to URL option enables you to back up your databases to Azure Blob Storage service. Azure Backup for SQL Server Virtual Machines provides a comprehensive enterprise backup solution that automatically manages your backups across your entire infrastructure.

Deployment options

All resources in Azure are managed and deployed through a common provider known as Azure Resource Manager. While there are various methods to deploy Azure resources, they ultimately converge into JSON documents called Azure Resource Manager templates, which serve as one of the deployment options for Azure resources.

Overview of Azure storage

Azure offers a fully redundant object-based storage model, and there are a few things to be aware of in designing and deploying Virtual Machines architecture. In terms of virtual machines, there are four types of storage you can use:

• Standard
• Standard SSD
• Premium SSD
• Ultra Disk

For production SQL Server data and transaction log files, you should only use Premium SSD storage and Ultra Disk. With premium storage, you see latencies in the range of 5-10 ms on a properly configured system. Alternatively, with Ultra Disk you may have sub millisecond latency but will likely see 1-2 ms workloads in the real world. You can use Standard storage for your database backups, as the performance is adequate for most backup and restore workloads.

High availability in Azure

The Azure platform is designed to be fault tolerant and provides quick recovery from service disruptions and transient errors. In fact, many organizations see higher levels of availability in single virtual machines deployments than they previously experienced in their on-premises environments. Microsoft guarantees uptime of at least 99.9% for single instance Azure virtual machine, when using Premium SSD or Ultra Disk for all disks.

Azure SQL Database is a Platform as a Service (PaaS) that provides high scalability capabilities, and it can be a great solution for certain workloads, and requires minimal maintenance efforts.

Purchasing model

SQL Database comes in two main purchasing models: vCore-based model and DTU-based model. Each purchasing model offers the following service tiers:

vCore-based

In this purchasing model, compute and storage resources are decoupled. It means you can scale storage and compute resources independently from one another. Here are listed the service tiers available:

Service tier	Capability
General Purpose	This service tier is designed for less intensive operations, and offers budget oriented balanced compute and storage options. It provides both provisioned compute tier and server less compute tier.
Business Critical	This service tier supports In-Memory OLTP, and built-in read-only replica. It also includes more memory per core, and uses local SSD storage, which is designed for performance-sensitive workloads.
Hyperscale	Hyperscale introduces horizontal scaling features that use advanced techniques to add compute nodes as the data sizes grow. It's only supported on single SQL database. Hyperscale allows you to scale storage and compute resources significantly over the limits available for the General Purpose and Business Critical service tiers.

DTU-based

In the DTU model, there are three service tiers available: Basic, Standard, and Premium. Compute and storage resources are dependent on the DTU level, and they provide a range of performance capabilities at a fixed storage limit, backup retention, and cost.

For example, if your database grows to the point it reaches the maximum storage limit, you would need to increase your DTU capacity, even if the compute utilization is low.

Microsoft Intelligent Data Platform roles

 

Role	Definition
Azure Data Engineer	Design and implement the management, monitoring, security, and privacy of data using the full stack of Azure data services to satisfy business needs.
Azure Data Analyst	Enable businesses to maximize the value of their data assets by using Microsoft Power BI. As a subject matter expert, Data Analysts are responsible for designing and building scalable data models, cleaning and transforming data, and enabling advanced analytic capabilities that provide meaningful business value through easy-to-comprehend data visualizations.
Azure Data Scientist	Applies their knowledge of data science and machine learning to implement and run machine learning workloads on Azure; in particular, using Azure Machine Learning Service.
Azure Artificial Intelligence Engineer	Use Cognitive Services, Machine Learning, and Knowledge Mining to architect and implement Microsoft AI solutions involving natural language processing, speech, computer vision, bots, and agents.
Azure Database Administrator	Implements and manages the operational aspects of cloud-native and hybrid data platform solutions built on Microsoft Azure data services and Microsoft SQL Server. The Azure Database Administrator uses a variety of methods and tools to perform day-to-day operations, including applying knowledge of using T-SQL for administrative management purposes.

Transformer

 Transformer models are used to solve all kinds of tasks across different modalities, including natural language processing (NLP), computer vision, audio processing, and more

The most basic object in the 🤗 Transformers library is the pipeline() function. It connects a model with its necessary preprocessing and postprocessing steps, allowing us to directly input any text and get an intelligible answer:

from transformers import pipeline

classifier = pipeline("sentiment-analysis")
classifier("I've been waiting for a HuggingFace course my whole life.")

[{'label': 'POSITIVE', 'score': 0.9598047137260437}]

By default, this pipeline selects a particular pretrained model that has been fine-tuned for sentiment analysis in English. The model is downloaded and cached when you create the classifier object. If you rerun the command, the cached model will be used instead and there is no need to download the model again.
There are three main steps involved when you pass some text to a pipeline:
The text is preprocessed into a format the model can understand.
The preprocessed inputs are passed to the model.
The predictions of the model are post-processed, so you can make sense of them
The pipeline() function supports multiple modalities, allowing you to work with text, images, audio, and even multimodal tasks

Text pipelines
text-generation: Generate text from a prompt
text-classification: Classify text into predefined categories
summarization: Create a shorter version of a text while preserving key information
translation: Translate text from one language to another
zero-shot-classification: Classify text without prior training on specific labels
feature-extraction: Extract vector representations of text
Image pipelines
image-to-text: Generate text descriptions of images
image-classification: Identify objects in an image
object-detection: Locate and identify objects in images
Audio pipelines
automatic-speech-recognition: Convert speech to text
audio-classification: Classify audio into categories
text-to-speech: Convert text to spoken audio
Multimodal pipelines
image-text-to-text: Respond to an image based on a text prompt
This is a common scenario in real-world projects because annotating text is usually time-consuming and requires domain expertise. For this use case, the zero-shot-classification pipeline is very powerful: it allows you to specify which labels to use for the classification, so you don’t have to rely on the labels of the pretrained model. You’ve already seen how the model can classify a sentence as positive or negative using those two labels — but it can also classify the text using any other set of labels you like.
from transformers import pipeline

classifier = pipeline("zero-shot-classification")
classifier(
    "This is a course about the Transformers library",
    candidate_labels=["education", "politics", "business"],
)
{'sequence': 'This is a course about the Transformers library',
 'labels': ['education', 'business', 'politics'],
 'scores': [0.8445963859558105, 0.111976258456707, 0.043427448719739914]}
This pipeline is called zero-shot because you don’t need to fine-tune the model on your data to use it. It can directly return probability scores for any list of labels you want!
Text generation
Now let’s see how to use a pipeline to generate some text. The main idea here is that you provide a prompt and the model will auto-complete it by generating the remaining text. This is similar to the predictive text feature that is found on many phones. Text generation involves randomness, so it’s normal if you don’t get the same results as shown below.
from transformers import pipeline

generator = pipeline("text-generation")
generator("In this course, we will teach you how to")
[{'generated_text': 'In this course, we will teach you how to understand and use '
                    'data flow and data interchange when handling user data. We '
                    'will be working with one or more of the most commonly used '
                    'data flows — data flows of various types, as seen by the '
                    'HTTP'}]
You can control how many different sequences are generated with the argument num_return_sequences and the total length of the output text with the argument max_length.

LLM 1

 LLMs are characterized by:

• Scale: They contain millions, billions, or even hundreds of billions of parameters
• General capabilities: They can perform multiple tasks without task-specific training
• In-context learning: They can learn from examples provided in the prompt
• Emergent abilities: As these models grow in size, they demonstrate capabilities that weren’t explicitly programmed or anticipated

The advent of LLMs has shifted the paradigm from building specialized models for specific NLP tasks to using a single, large model that can be prompted or fine-tuned to address a wide range of language tasks. This has made sophisticated language processing more accessible while also introducing new challenges in areas like efficiency, ethics, and deployment.

However, LLMs also have important limitations:

• Hallucinations: They can generate incorrect information confidently
• Lack of true understanding: They lack true understanding of the world and operate purely on statistical patterns
• Bias: They may reproduce biases present in their training data or inputs.
• Context windows: They have limited context windows (though this is improving)
• Computational resources: They require significant computational resources

Why is language processing challenging?

Computers don’t process information in the same way as humans. For example, when we read the sentence “I am hungry,” we can easily understand its meaning. Similarly, given two sentences such as “I am hungry” and “I am sad,” we’re able to easily determine how similar they are. For machine learning (ML) models, such tasks are more difficult. The text needs to be processed in a way that enables the model to learn from it. 

Even with the advances in LLMs, many fundamental challenges remain. These include understanding ambiguity, cultural context, sarcasm, and humor. LLMs address these challenges through massive training on diverse datasets, but still often fall short of human-level understanding in many complex scenarios.

LLM Course - Large Language Models

Understanding NLP and LLMs

What’s the difference?

• NLP (Natural Language Processing) is the broader field focused on enabling computers to understand, interpret, and generate human language. NLP encompasses many techniques and tasks such as sentiment analysis, named entity recognition, and machine translation.
• LLMs (Large Language Models) are a powerful subset of NLP models characterized by their massive size, extensive training data, and ability to perform a wide range of language tasks with minimal task-specific training. Models like the Llama, GPT, or Claude series are examples of LLMs that have revolutionized what’s possible in NLP

NLP is a field of linguistics and machine learning focused on understanding everything related to human language. The aim of NLP tasks is not only to understand single words individually, but to be able to understand the context of those words.

The following is a list of common NLP tasks, with some examples of each:

• Classifying whole sentences: Getting the sentiment of a review, detecting if an email is spam, determining if a sentence is grammatically correct or whether two sentences are logically related or not
• Classifying each word in a sentence: Identifying the grammatical components of a sentence (noun, verb, adjective), or the named entities (person, location, organization)
• Generating text content: Completing a prompt with auto-generated text, filling in the blanks in a text with masked words
• Extracting an answer from a text: Given a question and a context, extracting the answer to the question based on the information provided in the context
• Generating a new sentence from an input text: Translating a text into another language, summarizing a text

NLP isn’t limited to written text though. It also tackles complex challenges in speech recognition and computer vision, such as generating a transcript of an audio sample or a description of an image.

The Rise of Large Language Models (LLMs)

In recent years, the field of NLP has been revolutionized by Large Language Models (LLMs). These models, which include architectures like GPT (Generative Pre-trained Transformer)

A large language model (LLM) is an AI model trained on massive amounts of text data that can understand and generate human-like text, recognize patterns in language, and perform a wide variety of language tasks without task-specific training. They represent a significant advancement in the field of natural language processing (NLP).