Spaces:
Runtime error
Runtime error
File size: 1,337 Bytes
4a7b0b2 c91a3e7 8e1683e 4a7b0b2 8e1683e 4a7b0b2 8e1683e c91a3e7 4a7b0b2 8e1683e c91a3e7 4a7b0b2 c91a3e7 4a7b0b2 8e1683e 4a7b0b2 8e1683e c91a3e7 a0e9621 8e1683e 4a7b0b2 973f818 8e1683e c91a3e7 4a7b0b2 389147e |
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 |
import os
import gradio as gr
import torch
import PIL
from transformers import AutoProcessor, AutoModelForCausalLM # Using AutoModel classes
EXAMPLES_DIR = 'examples'
DEFAULT_PROMPT = "<image>"
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
# Load model using AutoModel with trust_remote_code=True
model = AutoModelForCausalLM.from_pretrained('dhansmair/flamingo-mini', trust_remote_code=True)
model.to(device)
model.eval()
# Initialize processor without the `device` argument
processor = AutoProcessor.from_pretrained('dhansmair/flamingo-mini')
# Setup some example images
examples = []
if os.path.isdir(EXAMPLES_DIR):
for file in os.listdir(EXAMPLES_DIR):
path = EXAMPLES_DIR + "/" + file
examples.append([path, DEFAULT_PROMPT])
def predict_caption(image, prompt):
assert isinstance(prompt, str)
# Process the image using the model
caption = model.generate(
processor(images=image, prompt=prompt), # Pass processed inputs to the model
max_length=50
)
if isinstance(caption, list):
caption = caption[0]
return caption
iface = gr.Interface(
fn=predict_caption,
inputs=[gr.Image(type="pil"), gr.Textbox(value=DEFAULT_PROMPT, label="Prompt")],
examples=examples,
outputs="text"
)
iface.launch(debug=True)
|