You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: 03-CoreGenerativeAITechniques/02-retrieval-augmented-generation.md
+1-1
Original file line number
Diff line number
Diff line change
@@ -13,7 +13,7 @@ Retrieval Augmented Generation (RAG) is a technique used to augment the response
13
13
There are 2 main phases in a RAG architecture: **Retrieval** and **Generation**.
14
14
15
15
-**Retrieval**: When the user poses a prompt, the system employs a retrieval mechanism of some sort to gather information from an external knowledge store. The knowledge store could be a vector database or a document, amongst other things.
16
-
-**Generation**: The retrieved information is then used to augment the user's prompt. This AI model processes both the retrieved info and the user's prompt to product an enriched response.
16
+
-**Generation**: The retrieved information is then used to augment the user's prompt. This AI model processes both the retrieved info and the user's prompt to produce an enriched response.
Copy file name to clipboardExpand all lines: 03-CoreGenerativeAITechniques/03-vision-audio.md
+4-4
Original file line number
Diff line number
Diff line change
@@ -10,13 +10,13 @@ In this lesson learn how vision AI allows your apps to generate and interpret im
10
10
11
11
_⬆️Click the image to watch the video⬆️_
12
12
13
-
Vision-based AI approaches are used to generate and interpret images. This can useful for a wide range of applications, such as image recognition, image generation, and image manipulation. Current models are multimodal, meaning they can accept a variety of inputs, such as text, images, and audio, and generate a variety of outputs. In this case, we are going to focus on image recognition.
13
+
Vision-based AI approaches are used to generate and interpret images. This can be useful for a wide range of applications, such as image recognition, image generation, and image manipulation. Current models are multimodal, meaning they can accept a variety of inputs, such as text, images, and audio, and generate a variety of outputs. In this case, we are going to focus on image recognition.
14
14
15
15
### Image recognition with MEAI
16
16
17
17
Image recognition is more than having the AI model tell you what it thinks is present in an image. You can also ask questions about the image, for example: _How many people are present and is it raining?_
18
18
19
-
Ok - so we're going to put the model through it's paces and ask it if it can tell us how many red shoes are in the first photo and then have it analyze a receipt that's in German so we know how much to tip.
19
+
Ok - so we're going to put the model through its paces and ask it if it can tell us how many red shoes are in the first photo and then have it analyze a receipt that's in German so we know how much to tip.
20
20
21
21

22
22
@@ -143,7 +143,7 @@ You'll need the **Microsoft.CognitiveServices.Speech** NuGet package for this ex
@@ -159,4 +159,4 @@ We have another example that [demonstrates how to perform real-time audio conver
159
159
160
160
You've learned how to add vision and audio capabilities to your .NET applications, in the next lesson find out how to create AI that has some ability to act autonomously.
Copy file name to clipboardExpand all lines: 03-CoreGenerativeAITechniques/04-agents.md
+4-4
Original file line number
Diff line number
Diff line change
@@ -45,7 +45,7 @@ Let's see a simple example of this in action!
45
45
46
46
### Build a math agent
47
47
48
-
We'll be building a single purpose agent that acts as a tutor to math students. It's sole purpose in life is to solve and then explain math problems the user asks.
48
+
We'll be building a single purpose agent that acts as a tutor to math students. Its sole purpose in life is to solve and then explain math problems the user asks.
49
49
50
50
1. To start with, we need to create an `AgentsClient` object that is responsible for managing the connection to Azure, the agent itself, the threads, the messages, and so on.
51
51
@@ -58,7 +58,7 @@ We'll be building a single purpose agent that acts as a tutor to math students.
0 commit comments