[Bugfix] fix chunker config bug (#1024)

Co-authored-by: Deven Patel <deven298@yahoo.com>
This commit is contained in:
Deven Patel
2023-12-18 17:32:26 +05:30
committed by GitHub
parent bee4e834b1
commit db37b2ac15
3 changed files with 21 additions and 17 deletions

View File

@@ -53,6 +53,7 @@ chunker:
chunk_size: 2000 chunk_size: 2000
chunk_overlap: 100 chunk_overlap: 100
length_function: 'len' length_function: 'len'
min_chunk_size: 0
``` ```
```json config.json ```json config.json
@@ -91,7 +92,8 @@ chunker:
"chunker": { "chunker": {
"chunk_size": 2000, "chunk_size": 2000,
"chunk_overlap": 100, "chunk_overlap": 100,
"length_function": "len" "length_function": "len",
"min_chunk_size": 0
} }
} }
``` ```
@@ -138,7 +140,8 @@ config = {
'chunker': { 'chunker': {
'chunk_size': 2000, 'chunk_size': 2000,
'chunk_overlap': 100, 'chunk_overlap': 100,
'length_function': 'len' 'length_function': 'len',
'min_chunk_size': 0
} }
} }
``` ```

View File

@@ -428,6 +428,7 @@ def validate_config(config_data):
Optional("chunk_size"): int, Optional("chunk_size"): int,
Optional("chunk_overlap"): int, Optional("chunk_overlap"): int,
Optional("length_function"): str, Optional("length_function"): str,
Optional("min_chunk_size"): int,
}, },
} }
) )

View File

@@ -1,24 +1,24 @@
import chainlit as cl
from embedchain import Pipeline as App
import os import os
import chainlit as cl
from embedchain import Pipeline as App
os.environ["OPENAI_API_KEY"] = "sk-xxx" os.environ["OPENAI_API_KEY"] = "sk-xxx"
@cl.on_chat_start @cl.on_chat_start
async def on_chat_start(): async def on_chat_start():
app = App.from_config(config={ app = App.from_config(
'app': { config={
'config': { "app": {"config": {"name": "chainlit-app"}},
'name': 'chainlit-app' "llm": {
} "config": {
}, "stream": True,
'llm': { }
'config': { },
'stream': True,
}
} }
}) )
# import your data here # import your data here
app.add("https://www.forbes.com/profile/elon-musk/") app.add("https://www.forbes.com/profile/elon-musk/")
app.collect_metrics = False app.collect_metrics = False
@@ -31,5 +31,5 @@ async def on_message(message: cl.Message):
msg = cl.Message(content="") msg = cl.Message(content="")
for chunk in await cl.make_async(app.chat)(message.content): for chunk in await cl.make_async(app.chat)(message.content):
await msg.stream_token(chunk) await msg.stream_token(chunk)
await msg.send() await msg.send()