Spaces:
Runtime error
Runtime error
File size: 2,342 Bytes
702d337 40de5f7 702d337 ebd223e 7d5edad 702d337 8921743 258046b 5c61a9c 258046b 702d337 46d5540 702d337 8aad5c5 702d337 46d5540 5c61a9c 702d337 |
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 |
import streamlit as st
import random
import requests
import io
from PIL import Image
from transformers import AutoProcessor, Blip2ForConditionalGeneration
from peft import PeftModel, PeftConfig
import torch
model = None
processor = None
st.title("Image Captioner - Caption the images")
st.markdown("Link to the model - [Image-to-Caption-App on 🤗 Spaces](https://huggingface.co/spaces/Shrey23/Image-Captioning)")
class UI:
def __init__(self):
mod = Model()
mod.load_model()
def displayUI(self):
image = st.file_uploader(label = "Upload your image here",type=['png','jpg','jpeg'])
if image is not None:
input_image = Image.open(image) #read image
st.image(input_image) #display image
with st.spinner("🤖 AI is at Work! "):
device = "cuda" if torch.cuda.is_available() else "cpu"
print(1)
inputs = processor(images=input_image, return_tensors="pt").to(device, torch.float32)
print(2)
pixel_values = inputs.pixel_values
# print(3)
# pixel_values = pixel_values.to(torch.float32)
print(4)
generated_ids = model.generate(pixel_values=pixel_values, max_length=25)
generated_caption = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
st.write(generated_caption)
st.success("Here you go!")
st.balloons()
else:
st.write("Upload an Image")
st.caption("NN Practical Project.")
class Model:
def load_model(self):
peft_model_id = "Shrey23/Image-Captioning"
config = PeftConfig.from_pretrained(peft_model_id)
global model
global processor
model = Blip2ForConditionalGeneration.from_pretrained(config.base_model_name_or_path, low_cpu_mem_usage=True, torch_dtype=torch.float32) #, device_map="auto", load_in_8bit=True
model = PeftModel.from_pretrained(model, peft_model_id)
processor = AutoProcessor.from_pretrained("Salesforce/blip2-opt-2.7b")
def main():
ui = UI()
ui.displayUI()
if __name__ == "__main__":
main()
|