Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update huggingface.py #425

Merged
merged 2 commits into from
Oct 5, 2023
Merged

Update huggingface.py #425

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
5b4e517
Update huggingface.py
Archit-Kohli Oct 5, 2023
7ce1dc9
Update huggingface.py add import torch statement
Archit-Kohli Oct 5, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
23 changes: 18 additions & 5 deletions application/llm/huggingface.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,13 +2,26 @@

class HuggingFaceLLM(BaseLLM):

def __init__(self, api_key, llm_name='Arc53/DocsGPT-7B'):
def __init__(self, api_key, llm_name='Arc53/DocsGPT-7B',q=False):
global hf

from langchain.llms import HuggingFacePipeline
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
tokenizer = AutoTokenizer.from_pretrained(llm_name)
model = AutoModelForCausalLM.from_pretrained(llm_name)
if q:
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline, BitsAndBytesConfig
tokenizer = AutoTokenizer.from_pretrained(llm_name)
bnb_config = BitsAndBytesConfig(

Check warning on line 13 in application/llm/huggingface.py

View check run for this annotation

Codecov / codecov/patch

application/llm/huggingface.py#L9-L13 

Added lines #L9 - L13 were not covered by tests
load_in_4bit=True,
bnb_4bit_use_double_quant=True,
bnb_4bit_quant_type="nf4",
bnb_4bit_compute_dtype=torch.bfloat16
)
model = AutoModelForCausalLM.from_pretrained(llm_name,quantization_config=bnb_config)

Check warning on line 19 in application/llm/huggingface.py

View check run for this annotation

Codecov / codecov/patch

application/llm/huggingface.py#L19 

Added line #L19 was not covered by tests
else:
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
tokenizer = AutoTokenizer.from_pretrained(llm_name)
model = AutoModelForCausalLM.from_pretrained(llm_name)

Check warning on line 23 in application/llm/huggingface.py

View check run for this annotation

Codecov / codecov/patch

application/llm/huggingface.py#L21-L23 

Added lines #L21 - L23 were not covered by tests

pipe = pipeline(
"text-generation", model=model,
tokenizer=tokenizer, max_new_tokens=2000,
Expand Down