File size: 981 Bytes
b5c0e0d | 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 | import streamlit as st
from transformers import AutoTokenizer, AutoModelForCausalLM
# Load the pre-trained model and tokenizer
model_name = "facebook/opt-125m"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)
st.title('Text Generation with Hugging Face')
# Input text
input_text = st.text_input("Enter your prompt:", "sun rises in the ")
if input_text:
# Tokenize the input text
inputs = tokenizer(input_text, return_tensors="pt")
# Generate text
outputs = model.generate(
inputs["input_ids"],
max_length=50, # Increased length for more content
num_return_sequences=1,
no_repeat_ngram_size=2,
top_p=0.95,
top_k=50,
temperature=0.7
)
# Decode the generated text
generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
# Display the generated text
st.write("Generated Text:")
st.write(generated_text)
|