Skip to content

Commit 7a2fd81

Browse files
authored
add link to cohere.ai
1 parent a508351 commit 7a2fd81

File tree

1 file changed

+3
-2
lines changed

1 file changed

+3
-2
lines changed

README.md

Lines changed: 3 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -127,7 +127,7 @@ Optionally, you can also install topically with BERTopic:
127127

128128
# How it works
129129

130-
Topically uses a generative language model (GPT) to assign a name to the text cluster. It sends a request to Cohere's managed model (get an [API key](https://dashboard.cohere.ai/welcome/register?utm_source=github&utm_medium=content&utm_campaign=sandbox&utm_content=topically) and use it for free for prototyping).
130+
Topically uses a generative language model (GPT) to assign a name to the text cluster. It sends a request to [Cohere](https://cohere.ai/)'s managed model (get an [API key](https://dashboard.cohere.ai/welcome/register?utm_source=github&utm_medium=content&utm_campaign=sandbox&utm_content=topically) and use it for free for prototyping).
131131

132132
To generate the titles, topically uses a couple of bundled prompts. To get the best names for your use case, it's best to edit the prompt to add more information about the context, and add good cluster names for 3-5 of your clusters.
133133

@@ -137,7 +137,8 @@ This works best on short texts (given the context length limitations of GPT mode
137137
Topically is pretty simple and early in its life. At the moment, it's made up of two main class:
138138

139139
### `Topically`
140-
This class maintains the client to the Cohere platform, and exposes the main interaction point with Topically (name_topics, at the moment). It lives in app.py.
140+
This class maintains the client to the
141+
platform, and exposes the main interaction point with Topically (name_topics, at the moment). It lives in app.py.
141142

142143
### `ClusterNamer`
143144
This class deals with preparing the prompts and calling the Generate endpoint to generate suggested topic names. It lives in cluster_namers.py.

0 commit comments

Comments
 (0)