Ubuntu
commited on
Commit
•
64a5bab
1
Parent(s):
66473f1
Updated Readme
Browse files
README
CHANGED
@@ -1,8 +1,39 @@
|
|
|
|
1 |
|
2 |
-
rasa
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
|
|
|
|
|
|
|
4 |
rasa run --credentials ./credentials.yml --enable-api --auth-token XYZ123 --model ./models --endpoints ./endpoints.yml --cors "*"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
|
6 |
-
|
|
|
|
|
|
|
7 |
|
8 |
-
Open localhost:8080 in browser.
|
|
|
1 |
+
# Readme
|
2 |
|
3 |
+
The source code in this repository shows a simple rasa chatbot fallback mechanism, which can get relavent information from Zir-AI search platform.
|
4 |
+
|
5 |
+
## Data
|
6 |
+
|
7 |
+
reviews.db contains the hotel reviews, its generated by using repo https://github.com/amin3141/zir-souffle
|
8 |
+
|
9 |
+
# How-to
|
10 |
+
|
11 |
+
First you need to have the rasa shell installed. You can do that by following the rasa tutorials https://rasa.com/docs/rasa/installation
|
12 |
+
|
13 |
+
Next is training the model
|
14 |
+
|
15 |
+
```
|
16 |
+
rasa train
|
17 |
+
```
|
18 |
|
19 |
+
Then to run in one terminal you need to run
|
20 |
+
|
21 |
+
```
|
22 |
rasa run --credentials ./credentials.yml --enable-api --auth-token XYZ123 --model ./models --endpoints ./endpoints.yml --cors "*"
|
23 |
+
```
|
24 |
+
|
25 |
+
And in the background or in another terminal run
|
26 |
+
|
27 |
+
```
|
28 |
+
rasa run actions
|
29 |
+
```
|
30 |
+
|
31 |
+
Lastly, to view the bot in the UI either simply load the `www/index.html` file in the browser or run `python3 server.py` and open `localhost:8080` in browser.
|
32 |
+
|
33 |
+
You could also just run the two commands and talk with the bot in shell
|
34 |
|
35 |
+
```
|
36 |
+
rasa run actions &
|
37 |
+
rasa shell
|
38 |
+
```
|
39 |
|
|