github/langchain
1
0
Fork 0
mirror of https://github.com/hwchase17/langchain.git synced 2025-08-12 14:23:58 +00:00
Code Issues Packages Projects Releases Wiki Activity

docs: model parameter mandatory for cohere embedding and rerank (#23349)

Browse Source 
Latest langchain-cohere sdk mandates passing in the model parameter into
the Embeddings and Reranker inits.

This PR is to update the docs to reflect these changes.
This commit is contained in:
Anirudh31415926535 2024-07-13 07:07:28 +08:00 committed by GitHub
parent aee55eda39
commit 7677ceea60
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
5 changed files with 15 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
docs/docs/how_to/embed_text.mdx
View File

@ -67,15 +67,16 @@ If you'd prefer not to set an environment variable you can pass the key in direc
```python ```python
from langchain_cohere import CohereEmbeddings from langchain_cohere import CohereEmbeddings
embeddings_model = CohereEmbeddings(cohere_api_key="...") embeddings_model = CohereEmbeddings(cohere_api_key="...", model='embed-english-v3.0')
``` ```
Otherwise you can initialize without any params: Otherwise you can initialize simply as shown below:
```python ```python
from langchain_cohere import CohereEmbeddings from langchain_cohere import CohereEmbeddings
embeddings_model = CohereEmbeddings() embeddings_model = CohereEmbeddings(model='embed-english-v3.0')
``` ```
Do note that it is mandatory to pass the model parameter while initializing the CohereEmbeddings class.
</TabItem> </TabItem>
<TabItem value="huggingface" label="Hugging Face"> <TabItem value="huggingface" label="Hugging Face">

11
docs/docs/integrations/retrievers/cohere-reranker.ipynb
View File

@ -309,9 +309,9 @@
"documents = TextLoader(\"../../how_to/state_of_the_union.txt\").load()\n", "documents = TextLoader(\"../../how_to/state_of_the_union.txt\").load()\n",
"text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=100)\n", "text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=100)\n",
"texts = text_splitter.split_documents(documents)\n", "texts = text_splitter.split_documents(documents)\n",
"retriever = FAISS.from_documents(texts, CohereEmbeddings()).as_retriever(\n", "retriever = FAISS.from_documents(\n",
" search_kwargs={\"k\": 20}\n", " texts, CohereEmbeddings(model=\"embed-english-v3.0\")\n",
")\n", ").as_retriever(search_kwargs={\"k\": 20})\n",
"\n", "\n",
"query = \"What did the president say about Ketanji Brown Jackson\"\n", "query = \"What did the president say about Ketanji Brown Jackson\"\n",
"docs = retriever.invoke(query)\n", "docs = retriever.invoke(query)\n",
@ -324,7 +324,8 @@
"metadata": {}, "metadata": {},
"source": [ "source": [
"## Doing reranking with CohereRerank\n", "## Doing reranking with CohereRerank\n",
"Now let's wrap our base retriever with a `ContextualCompressionRetriever`. We'll add an `CohereRerank`, uses the Cohere rerank endpoint to rerank the returned results." "Now let's wrap our base retriever with a `ContextualCompressionRetriever`. We'll add an `CohereRerank`, uses the Cohere rerank endpoint to rerank the returned results.\n",
"Do note that it is mandatory to specify the model name in CohereRerank!"
] ]
}, },
{ {
@ -339,7 +340,7 @@
"from langchain_community.llms import Cohere\n", "from langchain_community.llms import Cohere\n",
"\n", "\n",
"llm = Cohere(temperature=0)\n", "llm = Cohere(temperature=0)\n",
"compressor = CohereRerank()\n", "compressor = CohereRerank(model=\"rerank-english-v3.0\")\n",
"compression_retriever = ContextualCompressionRetriever(\n", "compression_retriever = ContextualCompressionRetriever(\n",
" base_compressor=compressor, base_retriever=retriever\n", " base_compressor=compressor, base_retriever=retriever\n",
")\n", ")\n",

4
docs/docs/integrations/text_embedding/cohere.ipynb
View File

@ -40,7 +40,9 @@
"metadata": {}, "metadata": {},
"outputs": [], "outputs": [],
"source": [ "source": [
"embeddings = CohereEmbeddings(model=\"embed-english-light-v3.0\")" "embeddings = CohereEmbeddings(\n",
" model=\"embed-english-light-v3.0\"\n",
") # It is mandatory to pass a model parameter to initialize the CohereEmbeddings object"
] ]
}, },
{ {

2
docs/docs/integrations/vectorstores/pgvector.ipynb
View File

@ -78,7 +78,7 @@
"# See docker command above to launch a postgres instance with pgvector enabled.\n", "# See docker command above to launch a postgres instance with pgvector enabled.\n",
"connection = \"postgresql+psycopg://langchain:langchain@localhost:6024/langchain\" # Uses psycopg3!\n", "connection = \"postgresql+psycopg://langchain:langchain@localhost:6024/langchain\" # Uses psycopg3!\n",
"collection_name = \"my_docs\"\n", "collection_name = \"my_docs\"\n",
"embeddings = CohereEmbeddings()\n", "embeddings = CohereEmbeddings(model=\"embed-english-v3.0\")\n",
"\n", "\n",
"vectorstore = PGVector(\n", "vectorstore = PGVector(\n",
" embeddings=embeddings,\n", " embeddings=embeddings,\n",

2
templates/cohere-librarian/cohere_librarian/blurb_matcher.py
View File

@ -23,7 +23,7 @@ parsed_data = [
] ]
parsed_data[1] parsed_data[1]
embeddings = CohereEmbeddings() embeddings = CohereEmbeddings(model="embed-english-v3.0")
docsearch = Chroma.from_texts( docsearch = Chroma.from_texts(
[x["title"] for x in parsed_data], embeddings, metadatas=parsed_data [x["title"] for x in parsed_data], embeddings, metadatas=parsed_data