Commit Graph

2305 Commits

Author SHA1 Message Date
Alpri Else
784d24a1d5
Support S3 Object keys with / in S3FileLoader (#1517)
Resolves https://github.com/hwchase17/langchain/issues/1510

### Problem
When loading S3 Objects with `/` in the object key (eg.
`folder/some-document.txt`) using `S3FileLoader`, the objects are
downloaded into a temporary directory and saved as a file.

This errors out when the parent directory does not exist within the
temporary directory.

See
https://github.com/hwchase17/langchain/issues/1510#issuecomment-1459583696
on how to reproduce this bug

### What this pr does
Creates parent directories based on object key. 

This also works with deeply nested keys:
`folder/subfolder/some-document.txt`
2023-03-08 16:17:26 -08:00
Harrison Chase
aba58e9e2e
Harrison/bumpver104 (#1525) 2023-03-08 09:46:02 -08:00
Harrison Chase
c4a557bdd4
add concept of prompt collection (#1507) 2023-03-08 08:31:29 -08:00
Ivan
97e3666e0d
changed requests.run to requests.get (#1485)
This pull request proposes an update to the Lightweight wrapper
library's documentation. The current documentation provides an example
of how to use the library's requests.run method, as follows:
requests.run("https://www.google.com"). However, this example does not
work for the 0.0.102 version of the library.

Testing:

The changes have been tested locally to ensure they are working as
intended.

Thank you for considering this pull request.
2023-03-07 21:10:23 -08:00
Harrison Chase
7ade419a0e
allow passing of messages into prompt template (#1505) 2023-03-07 21:10:12 -08:00
Harrison Chase
a4a2d79087
Harrison/rtd loader (#1513)
Co-authored-by: Youssef A. Abukwaik <yousseb@users.noreply.github.com>
2023-03-07 21:09:54 -08:00
Harrison Chase
8f21605d71
add return source docs (#1515) 2023-03-07 21:09:36 -08:00
Harrison Chase
064741db58
Harrison/fix text splitter (#1511)
Co-authored-by: ajaysolanky <ajsolanky@gmail.com>
Co-authored-by: Ajay Solanky <ajaysolanky@saw-l14668307kd.myfiosgateway.com>
2023-03-07 15:42:28 -08:00
Tom Dyson
e3354404ad
Fix link to Pinecone notebook (#1492) 2023-03-07 15:24:03 -08:00
Harrison Chase
3610ef2830
add fake embeddings class (#1503) 2023-03-07 15:23:46 -08:00
Ankush Gola
27104d4921
fix ChatOpenAI.agenerate (#1504) 2023-03-07 15:22:05 -08:00
Harrison Chase
4f41e20f09
memory docs (#1501) 2023-03-07 11:02:46 -08:00
Harrison Chase
d0062c7a9a
bump version to 103 (#1498) 2023-03-07 10:08:01 -08:00
Harrison Chase
8e6f599822
change to baselanguagemodel (#1496) 2023-03-07 09:29:59 -08:00
Harrison Chase
f276bfad8e
Harrison/chat memory (#1495) 2023-03-07 09:02:40 -08:00
Harrison Chase
7bec461782
Harrison/memory refactor (#1478)
moves memory to own module, factors out common stuff
2023-03-07 07:59:37 -08:00
kahkeng
df6865cd52
Allow no token limit for ChatGPT API (#1481)
The endpoint default is inf if we don't specify max_tokens, so unlike
regular completion API, we don't need to calculate this based on the
prompt.
2023-03-06 13:18:55 -08:00
Harrison Chase
312c319d8b
bump version to 102 (#1471) 2023-03-06 10:50:44 -08:00
Harrison Chase
0e21463f07
(rfc) chat models (#1424)
Co-authored-by: Ankush Gola <ankush.gola@gmail.com>
2023-03-06 08:34:24 -08:00
Juanky Soriano
dec3750875
Change method to calculate number of tokens for OpenAIChat (#1457)
Solves https://github.com/hwchase17/langchain/issues/1412

Currently `OpenAIChat` inherits the way it calculates the number of
tokens, `get_num_token`, from `BaseLLM`.
In the other hand `OpenAI` inherits from `BaseOpenAI`. 

`BaseOpenAI` and `BaseLLM` uses different methodologies for doing this.
The first relies on `tiktoken` while the second on `GPT2TokenizerFast`.

The motivation of this PR is:

1. Bring consistency about the way of calculating number of tokens
`get_num_token` to the `OpenAI` family, regardless of `Chat` vs `non
Chat` scenarios.
2. Give preference to the `tiktoken` method as it's serverless friendly.
It doesn't require downloading models which might make it incompatible
with `readonly` filesystems.
2023-03-06 07:20:25 -08:00
Tim Asp
763f879536
fix always verbose on summarization checker (#1440) 2023-03-05 07:10:08 -08:00
Harrison Chase
56b850648f
cr (#1436) 2023-03-04 08:38:56 -08:00
Harrison Chase
63a5614d23
Harrison/simple memory (#1435)
Co-authored-by: Tim Asp <707699+timothyasp@users.noreply.github.com>
2023-03-04 08:15:52 -08:00
Harrison Chase
a1b9dfc099
Harrison/similarity search chroma (#1434)
Co-authored-by: shibuiwilliam <shibuiyusuke@gmail.com>
2023-03-04 08:10:15 -08:00
Peng Qu
68ce68f290
Fix an unusual issue that occurs when using OpenAIChat for llm_math (#1410)
Fix an issue that occurs when using OpenAIChat for llm_math, refer to
the code style of the "Final Answer:" in Mrkl。 the reason is I found a
issue when I try OpenAIChat for llm_math, when I try the question in
Chinese, the model generate the format like "\n\nQuestion: What is the
square of 29?\nAnswer: 841", it translate the question first , then
answer. below is my snapshot:
<img width="945" alt="snapshot"
src="https://user-images.githubusercontent.com/82029664/222642193-10ecca77-db7b-4759-bc46-32a8f8ddc48f.png">
2023-03-04 07:56:07 -08:00
Ikko Eltociear Ashimine
b8a7828d1f
Update huggingface_datasets.ipynb (#1417)
HuggingFace -> Hugging Face
2023-03-04 00:22:31 -08:00
Kentaro Tanaka
6a4ee07e4f
Fix type hint of 'vectorstore_cls' arg in SemanticSimilarityExampleSelector (#1427)
Hello! Thank you for the amazing library you've created!

While following the tutorial at [the link(`Using an example
selector`)](https://langchain.readthedocs.io/en/latest/modules/prompts/examples/few_shot_examples.html#using-an-example-selector),
I noticed that passing Chroma as an argument to from_examples results in
a type hint error.

Error message(mypy):
```
Argument 3 to "from_examples" of "SemanticSimilarityExampleSelector" has incompatible type "Type[Chroma]"; expected "VectorStore"  [arg-type]mypy(error)
```

This pull request fixes the type hint and allows the VectorStore class
to be specified as an argument.
2023-03-04 00:20:18 -08:00
Tim Asp
23231d65a9
Add PyMuPDF PDF loader (#1426)
Different PDF libraries have different strengths and weaknesses. PyMuPDF
does a good job at extracting the most amount of content from the doc,
regardless of the source quality, extremely fast (especially compared to
Unstructured).

https://pymupdf.readthedocs.io/en/latest/index.html
2023-03-03 20:59:28 -08:00
3d54b05863
searx: add install instructions, update doc and notebooks (#1420)
- Added instructions on setting up self hosted searx
- Add notebook example with agent
- Use `localhost:8888` as example url to stay consistent since public
instances are not really usable.

Co-authored-by: blob42 <spike@w530>
2023-03-03 20:57:50 -08:00
Tim Asp
bca0935d90
[docs] fix minor import error (#1425) 2023-03-03 16:10:07 -08:00
Jon Luo
882f7964fb
fix sql misinterpretation of % in query (#1408)
% is being misinterpreted by sqlalchemy as parameter passing, so any
`LIKE 'asdf%'` will result in a value error with mysql, mariadb, and
maybe some others. This is one way to fix it - the alternative is to
simply double up %, like `LIKE 'asdf%%'` but this seemed cleaner in
terms of output.
Fixes #1383
2023-03-02 16:03:16 -08:00
JonLuca De Caro
443992c4d5
[Docs] Add missing word from prompt docs (#1406)
The prompt in the first example of the quickstart guide was missing `for
`
2023-03-02 16:02:54 -08:00
Eugene Yurtsev
a83a371069
Minor documentation update in initialize_agent (#1397)
Updating documentation in initialize_agent.

One thing that could benefit from further clarification is the
responsibility
breakdown by between an AgentExecutor vs. an Agent. The documentation
for an
AgentExecutor does not clarify that. From the class attributes, it
appears that
executor has access to the tools, while the agent is only aware of the
tool
names. Anyway, additional clarification would be beneficial on the
AgentExecutor class.
2023-03-02 11:46:35 -08:00
Nuno Campos
499e76b199
Allow the regular openai class to be used for ChatGPT models (#1393)
Co-authored-by: Harrison Chase <hw.chase.17@gmail.com>
2023-03-02 09:04:18 -08:00
Kacper Łukawski
8947797250
Return Cohere embeddings as lists of floats (#1394)
This PR fixes the types returned by Cohere embeddings. Currently, Cohere
client returns instances of `cohere.embeddings.Embeddings`. Since the
transport layer relies on JSON, some numbers might be represented as
ints, not floats, which happens quite often. While that doesn't seem to
be an issue, it breaks some pydantic models if they require strict
floats.
2023-03-02 09:02:10 -08:00
Jason Gill
1989e7d4c2
Update examples to prevent confusing missing _type warning (#1391)
The YAML and JSON examples of prompt serialization now give a strange
`No '_type' key found, defaulting to 'prompt'` message when you try to
run them yourself or copy the format of the files. The reason for this
harmless warning is that the _type key was not in the config files,
which means they are parsed as a standard prompt.

This could be confusing to new users (like it was confusing to me after
upgrading from 0.0.85 to 0.0.86+ for my few_shot prompts that needed a
_type added to the example_prompt config), so this update includes the
_type key just for clarity.

Obviously this is not critical as the warning is harmless, but it could
be confusing to track down or be interpreted as an error by a new user,
so this update should resolve that.
2023-03-02 07:39:57 -08:00
Harrison Chase
dda5259f68
bump version to 0.0.99 (#1390) 2023-03-02 07:25:59 -08:00
Kacper Łukawski
f032609f8d
Add recursive parameter to DirectoryLoader (#1389)
This PR allows loading a directory recursively.
2023-03-02 07:06:26 -08:00
Kacper Łukawski
9ac442624c
Add Qdrant named arguments (#1386)
This PR:
- Increases `qdrant-client` version to 1.0.4
- Introduces custom content and metadata keys (as requested in #1087)
- Moves all the `QdrantClient` parameters into the method parameters to
simplify code completion
2023-03-02 07:05:14 -08:00
Francisco Ingham
34abcd31b9
remove limit clause from prompt for compatibility with ms sql server (#1385)
For reference see:
8a35811556

Co-authored-by: Francisco Ingham <>
2023-03-02 07:02:42 -08:00
Ankush Gola
fe30be6fba
add async and streaming support to OpenAIChat (#1378)
title says it all
2023-03-01 21:55:43 -08:00
Lakshya Agarwal
cfed0497ac
Minor grammatical fixes (#1325)
Fixed typos and links in a few places across documents
2023-03-01 21:18:09 -08:00
Ryan Dao
59157b6891
Bug: Fix Python version validation in PythonAstREPLTool (#1373)
The current logic checks if the Python major version is < 8, which is
wrong. This checks if the major and minor version is < 3.9.
2023-03-01 21:15:27 -08:00
Harrison Chase
e178008b75
Harrison/track token usage (#1382)
Co-authored-by: Zak King <zaking17@gmail.com>
2023-03-01 21:15:13 -08:00
Harrison Chase
1cd8996074
Harrison/summarizer chain (#1356)
Co-authored-by: Tim Asp <707699+timothyasp@users.noreply.github.com>
2023-03-01 20:59:07 -08:00
yakigac
cfae03042d
Fix the openaichat example (#1377)
The example was wrong.
2023-03-01 18:24:32 -08:00
Harrison Chase
4b5e850361
chatgpt wrapper (#1367) 2023-03-01 11:47:01 -08:00
Harrison Chase
4d4b43cf5a
fix doc names (#1354) 2023-03-01 09:40:31 -08:00
Harrison Chase
c01f9100e4
bump version to 0097 (#1365) 2023-03-01 08:20:24 -08:00
Christie Jacob
edb3915ee7
typo in vectorstores (#1362) 2023-03-01 07:21:37 -08:00