Spaces:
Sleeping
Sleeping
import gradio as gr | |
from transformers import AutoProcessor, AutoTokenizer, AutoImageProcessor, AutoModelForCausalLM, BlipForConditionalGeneration, VisionEncoderDecoderModel | |
import torch | |
git_processor_base = AutoProcessor.from_pretrained("microsoft/git-base-coco") | |
git_model_base = AutoModelForCausalLM.from_pretrained("microsoft/git-base-coco") | |
git_processor_large = AutoProcessor.from_pretrained("microsoft/git-large-coco") | |
git_model_large = AutoModelForCausalLM.from_pretrained("microsoft/git-large-coco") | |
blip_processor_base = AutoProcessor.from_pretrained("Salesforce/blip-image-captioning-base") | |
blip_model_base = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base") | |
blip_processor_large = AutoProcessor.from_pretrained("Salesforce/blip-image-captioning-large") | |
blip_model_large = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large") | |
vitgpt_processor = AutoImageProcessor.from_pretrained("nlpconnect/vit-gpt2-image-captioning") | |
vitgpt_model = VisionEncoderDecoderModel.from_pretrained("nlpconnect/vit-gpt2-image-captioning") | |
vitgpt_tokenizer = AutoTokenizer.from_pretrained("nlpconnect/vit-gpt2-image-captioning") | |
device = "cuda" if torch.cuda.is_available() else "cpu" | |
git_model_base.to(device) | |
blip_model_base.to(device) | |
git_model_large.to(device) | |
blip_model_large.to(device) | |
vitgpt_model.to(device) | |
def generate_caption(processor, model, image, tokenizer=None): | |
inputs = processor(images=image, return_tensors="pt").to(device) | |
generated_ids = model.generate(pixel_values=inputs.pixel_values, max_length=50) | |
if tokenizer is not None: | |
generated_caption = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0] | |
else: | |
generated_caption = processor.batch_decode(generated_ids, skip_special_tokens=True)[0] | |
return generated_caption | |
def generate_captions(image): | |
caption_git_base = generate_caption(git_processor_base, git_model_base, image) | |
caption_git_large = generate_caption(git_processor_large, git_model_large, image) | |
caption_blip_base = generate_caption(blip_processor_base, blip_model_base, image) | |
caption_blip_large = generate_caption(blip_processor_large, blip_model_large, image) | |
caption_vitgpt = generate_caption(vitgpt_processor, vitgpt_model, image, vitgpt_tokenizer) | |
return caption_git_base, caption_git_large, caption_blip_base, caption_blip_large, caption_vitgpt | |
examples = [["cat.jpg"], ["dog.jpg"], ["horse.jpg"]] | |
outputs = [gr.outputs.Textbox(label="Caption generated by GIT-base"), gr.outputs.Textbox(label="Caption generated by GIT-large"), gr.outputs.Textbox(label="Caption generated by BLIP-base"), gr.outputs.Textbox(label="Caption generated by BLIP-large"), gr.outputs.Textbox(label="Caption generated by ViT+GPT-2")] | |
title = "Image to Text : Multiple Models" | |
description = "Explore the Gradio Demo for comparing three state-of-the-art vision+language models: GIT, BLIP, and ViT+GPT2. To use the demo, upload your image and click 'submit,' or choose from the provided examples." | |
article = "<p style='text-align: center'><a href='https://huggingface.co/docs/transformers/main/model_doc/blip' target='_blank'>BLIP docs</a> | <a href='https://huggingface.co/docs/transformers/main/model_doc/git' target='_blank'>GIT docs</a></p>" | |
iface = gr.Interface(fn=generate_captions, | |
inputs=gr.inputs.Image(type="pil"), | |
outputs=outputs, | |
examples=examples, | |
title=title, | |
description=description, | |
article=article, | |
enable_queue=True) | |
iface.launch(server_name="0.0.0.0", server_port=7860) | |
''' | |
import gradio as gr | |
import numpy as np | |
from PIL import Image | |
def generate_ascii_art(image): | |
try: | |
# Convert the numpy array to a PIL Image | |
img = Image.fromarray(np.uint8(image)) | |
# Resize the image to a smaller size for faster processing | |
img = img.resize((80, 60)) | |
# Convert the image to grayscale | |
img = img.convert("L") | |
# Define ASCII characters to represent different intensity levels | |
#ascii_chars = "@%#*+=-:. " | |
ascii_chars = "$@B%8&WM#*oahkbdpqwmZO0QLCJUYXzcvunxrjft/|()1{}[]?-_+~<>i!lI;:,\\^`'. " | |
# Convert each pixel to ASCII character based on intensity | |
ascii_image = "" | |
for pixel_value in img.getdata(): | |
ascii_image += ascii_chars[pixel_value // 25] | |
# Reshape the ASCII string to match the resized image dimensions | |
ascii_image = "\n".join([ascii_image[i:i + img.width] for i in range(0, len(ascii_image), img.width)]) | |
return ascii_image | |
except Exception as e: | |
return f"Error: {e}" | |
iface = gr.Interface( | |
fn=generate_ascii_art, | |
inputs="image", | |
outputs="text", | |
title="ASCII Art Generator", | |
description="Upload an image, and this app will turn it into ASCII art! - Simple Gradio App from Docker", | |
live=True | |
) | |
iface.launch(server_name="0.0.0.0", server_port=7860) | |
import gradio as gr | |
import subprocess | |
def run_command(command): | |
try: | |
result = subprocess.check_output(command, shell=True, text=True) | |
return result | |
except subprocess.CalledProcessError as e: | |
return f"Error: {e}" | |
iface = gr.Interface( | |
fn=run_command, | |
inputs="text", | |
outputs="text", | |
#live=True, | |
title="Command Output Viewer", | |
description="Enter a command and view its output.", | |
examples=[ | |
["ls"], | |
["pwd"], | |
["echo 'Hello, Gradio!'"]] | |
) | |
iface.launch(server_name="0.0.0.0", server_port=7860) | |
''' |