• GenAI Stack version: 0.2.0
• Python version: 3.8+

Hacktoberfest Accepted PR guidelines

• Please check the documentation and CONTRIBUTING.md before you start making changes
• We accept the PR as hacktoberfest-accepted for this issue, if the PR adds the feature of LLM Chain i.e., connecting two LLM model response and make a chain.

• GenAI Stack version: 2.0.5
• Python version: 3.8
• Operating System: macOS Monterey

Description

Currently ETL RestAPI doesn't have a pydantic data model for Request Body. RestAPI's are built using the FastAPI framework and the documentation is autogenerated by the FastAPI itself.

from the above image the submit-job endpoint which is a post method and there is no information about the request body structure and type. so adding a data model would give the information about the request body.

Here are the documentations related to
GenAI Stack Server : https://genaistack.aiplanet.com/advanced-guide/genai_stack_server
API's Reference : https://genaistack.aiplanet.com/advanced-guide/openapi

• GenAI Stack version: 0.2.0
• Python version: 3.8+

Hacktoberfest Accepted PR guidelines

• Please check the documentation and CONTRIBUTING.md before you start making changes
• We accept the PR as hacktoberfest-accepted for this issue, if the PR contains a use case with application that makes an impact

• GenAI Stack version:
• Python version:3.6
• Operating System:windows

Description

Trying to install required libs from ui/requiremts.txt but getting version related issues

ERROR: Could not find a version that satisfies the requirement attrs==23.1.0 (from -r requirements.txt (line 2)) (from versions: 15.0.0a1, 15.0.0, 15.1.0, 15.2.0, 16.0.0, 16.1.0, 16.2.0, 16.3.0, 17.1.0, 17.2.0, 17.3.0, 17.4.0, 18.1.0, 18.2.0, 19.1.0, 19.2.0, 19.3.0, 20.1.0, 20.2.0, 20.3.0, 21.1.0, 21.2.0, 21.3.0, 21.4.0, 22.1.0, 22.2.0)
ERROR: No matching distribution found for attrs==23.1.0 (from -r requirements.txt (line 2))

ERROR: Could not find a version that satisfies the requirement blinker==1.6.2 (from -r requirements.txt (line 3)) (from versions: 0.8, 0.9, 1.0, 1.1, 1.2, 1.3, 1.4, 1.5)
ERROR: No matching distribution found for blinker==1.6.2 (from -r requirements.txt (line 3))

What I Did

pip install -r requirements.txt

if we remove the version, it is installing

• GenAI Stack version: 0.2.0
• Python version: 3.8+

Hacktoberfest Accepted PR guidelines

• Please check the documentation and CONTRIBUTING.md before you start making changes
• We accept the PR as hacktoberfest-accepted for this issue, if the PR adds new model in the Model list
  • Currently we support GPT4All and GPT3.5, we appreciate adding new open source LLM model support

We need a functionality to use pipelines directly in the HuggingFace model. Lot of DataScientists are very well comfortable with declaring pipelines from Hugging face directly instead of passing it through model_kwargs and pipeline_kwargs which makes it too confusing for them:

How we are currently building pipelines:

llm = HuggingFaceModel.from_kwargs(model=model_name_or_path,
 model_kwargs={"device_map":"cuda","quantization_config":quantization_config,"trust_remote_code":"False","low_cpu_mem_usage":"True"},
 task='text-generation',
 pipeline_kwargs=({"max_new_tokens":512,"do_sample":"True","temperature":0.7,"top_p":0.95,"top_k":40,"repetition_penalty":1.1}))

How some data scientists are expecting to use hugging face:

from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline

model_name_or_path = "TheBloke/Mistral-7B-Instruct-v0.1-GPTQ"
model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
device_map="auto",
trust_remote_code=False,
revision="gptq-8bit-32g-actorder_True")

tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
pipe = pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
max_new_tokens=512,
do_sample=True,
temperature=0.7,
top_p=0.95,
top_k=40,
repetition_penalty=1.1
)
model = HuggingFaceModel.from_kwargs(pipeline = pipe)

While the latter needs more lines of code it gives much more control and customisability in the hands of the data scientist in declaring his model. We can add one more kwarg pipeline in which the user can specify the pipeline directly.

• GenAI Stack version: 0.2.0
• Python version: 3.8+

Hacktoberfest Accepted PR guidelines

• Please check the documentation and CONTRIBUTING.md before you start making changes
• We accept the PR as hacktoberfest-accepted for this issue, if the PR adds new blogs in the documentation.
  • You can write a blog on GenAI Stack in markdown. We will update it as Community Spotlight along with your name

• GenAI Stack version: 0.2.0
• Python version: 3.8+
• Operating System: Linux/Windows/Mac

Description

Add custom loaders rather than relying on other tools(Can be contributed individually).

• PDF
• CSV
• Text(.txt)
• Excel (xlsx)
• JSON
• Webpage
• Youtube
• Directory of Documents(specific extensions or all from the given directory)
• Jira
• hubspot
• airbyte
• Databases(MySQL/MariaDB, PostgreSQL, MongoDB, etc...)
• Git Repo

• GenAI Stack version: 0.2.0
• Python version: 3.8+

Hacktoberfest Accepted PR guidelines

• Please check the documentation and CONTRIBUTING.md before you start making changes
• We accept the PR as hacktoberfest-accepted for this issue, if the PR improves the existing documentation by addition flow chart diagrams to the model

• GenAI Stack version: 0.2.5
• Python version: 3.10
• Operating System: Ubuntu(WSL)

Description

In some cases even though there is a context present in the source with respect to the question, its unable to query it.

Question : What is the bottom up process?
Response generated : {'output': 'Based on the given context, there is no information provided about the bottom-up process. Therefore, it is not possible to answer the question.'} Response time : 1.0319557189941406

Source file - Employee-Stock-Option-Plans-ESOP-Best-Practices-2.pdf

• GenAI Stack version: 0.2.5
• Python version: 3.8
• Operating System: Ubuntu

Description

Query arguement in retriever.get_chat_history() seems a bit ambiguous. Hence it can be removed.

There are a couple of edge cases that we want to tackle in the caching layer.

Description

There is no CODE_OF_CONDUCT.md file in your repo, I will happy to add one
, if you assign me this under hacktoberfest

• GenAI Stack version: 0.2.0
• Python version: 3.8+

Hacktoberfest Accepted PR guidelines

• Please check the documentation and CONTRIBUTING.md before you start making changes
• We accept the PR as hacktoberfest-accepted for this issue, if the PR adds the support of YouTube Video loaders in ETL.

• GenAI Stack version: 0.2.0
• Python version: 3.8+

Hacktoberfest Accepted PR guidelines

• Please check the documentation and CONTRIBUTING.md before you start making changes
• We accept the PR as hacktoberfest-accepted for this issue, if the PR adds the support to any one of this Vector Database
  • Milvus
  • Qdrant
  • FAISS
• Please comment your name along with the Vector Database name to assign the contribution work to you