From the uDemy course on LLM engineering.
https://www.udemy.com/course/llm-engineering-master-ai-and-large-language-models
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
94 lines
2.8 KiB
94 lines
2.8 KiB
WEBVTT |
|
|
|
00:00.950 --> 00:02.300 |
|
And here we have it. |
|
|
|
00:02.300 --> 00:04.100 |
|
The user interface is completed. |
|
|
|
00:04.100 --> 00:11.150 |
|
The extra notification came through on my phone and my watch, and you can see the wonderful trace of |
|
|
|
00:11.150 --> 00:16.070 |
|
what went on here with the surfaced conclusion at the bottom. |
|
|
|
00:16.280 --> 00:22.010 |
|
And it's actually just you just saw it up in the memory and it's just going off right now because it's |
|
|
|
00:22.010 --> 00:24.560 |
|
kicking off another run while we speak. |
|
|
|
00:24.590 --> 00:27.830 |
|
Uh, I left it too long, but there you got the idea. |
|
|
|
00:27.830 --> 00:35.420 |
|
It ran, it completed, we saw the latest memories, and it was able to show the trace of the thinking |
|
|
|
00:35.420 --> 00:37.220 |
|
that it was doing and what was happening. |
|
|
|
00:37.220 --> 00:40.160 |
|
And it will be repeating this every five minutes. |
|
|
|
00:40.160 --> 00:41.990 |
|
I should probably slow that down a bit. |
|
|
|
00:42.110 --> 00:48.380 |
|
And notifying me all through the day and night of new deals. |
|
|
|
00:48.560 --> 00:54.470 |
|
So hopefully this has been, uh, satisfying for you to see it come together. |
|
|
|
00:54.470 --> 00:59.600 |
|
Hopefully you're running this yourself and seeing the same thing, seeing the conversation between the |
|
|
|
00:59.600 --> 01:05.870 |
|
different agents in our agent framework and seeing this being surfaced and highlighted in this user |
|
|
|
01:05.870 --> 01:13.250 |
|
interface, and seeing how easy, of course, how incredibly easy it is to use Gradio to build something |
|
|
|
01:13.280 --> 01:21.200 |
|
off the shelf that is able to give us such great insights into what's going on and give us so much functionality, |
|
|
|
01:21.200 --> 01:23.330 |
|
all in a set of screens. |
|
|
|
01:23.330 --> 01:28.700 |
|
So with this, I hope that you enjoy working with this agent framework. |
|
|
|
01:28.700 --> 01:29.960 |
|
Make it your own. |
|
|
|
01:29.990 --> 01:35.960 |
|
I will probably go and fix the fact that the screen should refresh whilst it's also processing, so |
|
|
|
01:35.960 --> 01:37.250 |
|
you don't have to to wait. |
|
|
|
01:37.250 --> 01:42.410 |
|
But if I don't do it, then by all means you should do it yourself and add on more to this user interface |
|
|
|
01:42.410 --> 01:43.820 |
|
and make this your own. |
|
|
|
01:43.820 --> 01:50.720 |
|
Not only have you hopefully trained your own LLM to be even more accurate than R1, but also you're |
|
|
|
01:50.720 --> 01:55.790 |
|
adding more functionality, improving the prompts, and finding other ways that we can bring this to |
|
|
|
01:55.820 --> 02:00.290 |
|
life and also add add more features and surface them in the UI. |
|
|
|
02:00.290 --> 02:06.200 |
|
And then when you've done that, push your code so that I can see it and admire it, and other students |
|
|
|
02:06.230 --> 02:08.780 |
|
can take advantage of the changes that you've made. |
|
|
|
02:08.990 --> 02:15.680 |
|
Enjoy working with this agent framework, and I will see you for the next video where we wrap things |
|
|
|
02:15.680 --> 02:16.220 |
|
up.
|
|
|