Spaces:
Sleeping
Sleeping
File size: 2,429 Bytes
37fedd6 4ce15d7 8b8242c 32414d2 37fedd6 4ce15d7 fc92fad 4ce15d7 37fedd6 bde561e 8b8242c b479b99 fc92fad b479b99 4ce15d7 6161368 a75db23 b479b99 6161368 8b8242c 6161368 a75db23 6161368 a75db23 8b8242c 6161368 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 |
# import os
# import streamlit as st
# import torch
# from transformers import AutoTokenizer, AutoModelForCausalLM
# from huggingface_hub import login
# # Load Hugging Face Token from Secrets
# hf_token = os.getenv("HF_TOKEN")
# if not hf_token:
# st.error("Hugging Face token is missing! Please add it to Hugging Face Secrets.")
# st.stop()
# # Authenticate
# login(token=hf_token)
# # Load the model and tokenizer with authentication
# MODEL_NAME = "google/gemma-2b-it"
# tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, token=hf_token)
# model = AutoModelForCausalLM.from_pretrained(MODEL_NAME, token=hf_token, torch_dtype=torch.float16, device_map="auto")
# # Streamlit UI
# st.title("Gemma-2B Code Assistant")
# user_input = st.text_area("Enter your coding query:")
# if st.button("Generate Code"):
# if user_input:
# inputs = tokenizer(user_input, return_tensors="pt").to("cuda")
# output = model.generate(**inputs, max_new_tokens=100)
# response = tokenizer.decode(output[0], skip_special_tokens=True)
# st.write(response)
# else:
# st.warning("Please enter a query!")
import os
import streamlit as st
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
# Load Hugging Face Token
hf_token = os.getenv("HF_TOKEN")
if not hf_token:
st.error("❌ Hugging Face token is missing! Please add it to Secrets.")
st.stop()
# Set device to CPU (because CUDA is unavailable)
device = "cpu"
# Load tokenizer and model in CPU mode (without bitsandbytes)
MODEL_NAME = "google/gemma-2b-it"
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, token=hf_token)
model = AutoModelForCausalLM.from_pretrained(
MODEL_NAME,
token=hf_token,
torch_dtype=torch.float32, # 👈 Use standard float32 for CPU
device_map="cpu"
)
# Streamlit UI
st.title("Gemma-2B Code Assistant")
user_input = st.text_area("Enter your coding query:")
if st.button("Generate Code"):
if user_input:
with st.spinner("⏳ Generating response... Please wait!"):
inputs = tokenizer(user_input, return_tensors="pt").to(device)
output = model.generate(**inputs, max_new_tokens=50)
response = tokenizer.decode(output[0], skip_special_tokens=True)
st.subheader("📝 Generated Code:")
st.code(response, language="python")
else:
st.warning("⚠️ Please enter a query!")
|