This repository contains hands-on code samples, datasets, and demo projects designed to support learning and exam preparation for AI-102: Designing and Implementing an Azure AI Solution certification.
Each module aligns with the core topics of the AI-102 exam, featuring practical implementations of Azure Cognitive Services and Azure Machine Learning components.
- Simple app using Azure OpenAI for prompt completion and text generation.
templates/,app.py, andrequirements.txtprovided for quick deployment.
- Demo project using CLU to understand user intents and extract entities.
- Includes a JSON file (
pizza_clu_sample_utterances.json) with training utterances.
- Custom NER implementation using labeled
.txtsamples fromfinancial_entities_txt. - Useful for training Azure Language Studio to recognize financial terms like stock tickers, forex, and cryptocurrencies.
- Question Answering app using custom knowledge bases.
- Built to integrate with Azure Language Services.
- Demonstrates how to create a custom skillset using:
index.json,indexer.json,skillset.json- Example files:
employee-report.txt,project-plan.txt,meeting-minutes.txt
- Uses
training_data_customto train a classification model in Azure Language Studio. - Classifies documents into predefined categories.
- Analyze US Green Card documents (
id1.jpeg,id2.jpg) using Azure Document Intelligence. app.pyimplements document analysis and field extraction.
- Azure Cognitive Search pipeline example to extract structured data from resumes.
- Includes PDF files and complete
index.json,indexer.json, andskillset.json.
- Detect faces from images using Azure Face Service.
app_face_detection.pyfor the Face API demo.
- Analyze images for objects, OCR, and tags using:
app_image_analysis.pyapp_ocr.py
- Scripts showcasing various Azure Language services:
app_entity_linking.pyapp_key_phrase_extraction.pyapp_lang_detection.pyapp_ner.pyapp_pii_detection.pyapp_sentiment_analysis.pyapp_summarization.py
- Example app for combining Azure OpenAI with Azure AI Search.
- PDF
Project_Orion_Confidential.pdfused for RAG implementation.
- Text-to-speech (
app_tts.py) - Speech-to-text (
app_stt.py) - Translation (
app_translate.py) - SSML custom voices (
app_ssml.py) - Output file:
output_audio.wav
- Translate text using Azure Translation API.