Update README.md
Browse files
README.md
CHANGED
@@ -2,3 +2,48 @@
|
|
2 |
pipeline_tag: text2text-generation
|
3 |
---
|
4 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
pipeline_tag: text2text-generation
|
3 |
---
|
4 |
|
5 |
+
# Aeona
|
6 |
+
|
7 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
8 |
+
![Aeona Banner](https://wwww.aeona.xyz/banner.png)
|
9 |
+
|
10 |
+
# Model Details
|
11 |
+
|
12 |
+
An generative AI based on [google flan base](https://huggingface.co/google/flan-t5-base).
|
13 |
+
|
14 |
+
Recommended to use along with an [AIML Chatbot](https://github.com/deepsarda/Aeona-Aiml) to reduce load, get better replies, add name and personality to your bot.
|
15 |
+
Using an AIML Chatbot will allow you to hardcode some replies also.
|
16 |
+
|
17 |
+
## Model Description
|
18 |
+
|
19 |
+
eona is an chatbot which hope's to be able to talk with humans as if its an friend!
|
20 |
+
It's main target platform is discord.
|
21 |
+
You can invite the bot [here](https://aeona.xyz).
|
22 |
+
|
23 |
+
To learn more about this project and chat with the ai, you can use this [website](https://aeona.xyx/).
|
24 |
+
|
25 |
+
Aeona works by sending a query to the AIML Chatbot if the AIML chabot is unable to answer then it comes over here to ***apply some real brains*** so to speak.
|
26 |
+
The main reason for using google flan is its great performance at logical reasoning.
|
27 |
+
# Participate and Help the AI improve or just hang out at [hugging face discussions](https://huggingface.co/deepparag/Aeona/discussions)
|
28 |
+
|
29 |
+
## Goals
|
30 |
+
The goal is to create an AI which will work with AIML in order to create the most human like AI.
|
31 |
+
|
32 |
+
#### Why not an AI on its own?
|
33 |
+
For AI it is not possible (realistically) to learn about the user and store data on them, when compared to an AIML which can even execute code!
|
34 |
+
The goal of the AI is to generate responses where the AIML fails.
|
35 |
+
|
36 |
+
Hence the goals becomes to make an AI which has a wide variety of knowledge, yet be as small as possible!
|
37 |
+
So we use 3 dataset:-
|
38 |
+
1. [Movielines](https://www.kaggle.com/Cornell-University/movie-dialog-corpus) The movie lines promote longer and more thought out responses but it can be very random. About 200k lines!
|
39 |
+
2. [Discord Messages](https://www.kaggle.com/jef1056/discord-data) The messages are on a wide variety of topics filtered and removed spam which makes the AI highly random but gives it a very random response to every days questions! about 120 million messages!
|
40 |
+
3. Custom dataset scrapped from my messages, These messages are very narrow teaching this dataset and sending a random reply will make the AI say sorry loads of time!
|
41 |
+
|
42 |
+
## Training
|
43 |
+
The Discord Messages Dataset simply dwarfs the other datasets, Hence the data sets are repeated.
|
44 |
+
This leads to them covering each others issues!
|
45 |
+
|
46 |
+
The AI has a context of 6 messages which means it will reply until the 4th message from user.
|
47 |
+
[Example](https://huggingface.co/deepparag/Aeona-Beta/discussions/1)
|
48 |
+
|
49 |
+
|