Spaces:
Running
Running
File size: 2,728 Bytes
fef0a8d 99eb93c fef0a8d 0c034e2 fef0a8d 5268082 fef0a8d 96f2f76 fef0a8d 5268082 201c325 5268082 |
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 76 77 78 |
# Imports
import gradio as gr
import spaces
import torch
from PIL import Image
from transformers import AutoModel, AutoTokenizer
# Pre-Initialize
DEVICE = "auto"
if DEVICE == "auto":
DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
print(f"[SYSTEM] | Using {DEVICE} type compute device.")
# Variables
DEFAULT_INPUT = "Describe in one paragraph."
repo = AutoModel.from_pretrained("openbmb/MiniCPM-V-2_6", torch_dtype=torch.bfloat16, trust_remote_code=True)
tokenizer = AutoTokenizer.from_pretrained("openbmb/MiniCPM-V-2_6", trust_remote_code=True)
css = '''
.gradio-container{max-width: 560px !important}
h1{text-align:center}
footer {
visibility: hidden
}
'''
# Functions
@spaces.GPU(duration=60)
def generate(image, instruction=DEFAULT_INPUT, sampling=False, temperature=0.7, top_p=0.8, top_k=100, repetition_penalty=1.05, max_tokens=512):
global model, tokenizer
image_rgb = Image.open(image).convert("RGB")
print(image_rgb, instruction)
inputs = [{"role": "user", "content": [image_rgb, instruction]}]
parameters = {
"sampling": sampling,
"temperature": temperature,
"top_p": top_p,
"top_k": top_k,
"repetition_penalty": repetition_penalty,
"max_new_tokens": max_tokens
}
output = model.chat(image=None, msgs=inputs, tokenizer=tokenizer, **parameters)
return output
def cloud():
print("[CLOUD] | Space maintained.")
# Initialize
with gr.Blocks(css=css) as main:
with gr.Column():
gr.Markdown("🪄 Analyze images and caption them using state-of-the-art openbmb/MiniCPM-V-2_6.")
with gr.Column():
input = gr.Image(label="Image")
instruction = gr.Textbox(lines=1, value=DEFAULT_INPUT, label="Instruction")
sampling = gr.Checkbox(value=False, label="Sampling")
temperature = gr.Slider(minimum=0, maximum=2, step=0.01, value=0.7, label="Temperature")
top_p = gr.Slider(minimum=0, maximum=1, step=0.01, value=0.8, label="Top P")
top_k = gr.Slider(minimum=0, maximum=1000, step=1, value=100, label="Top K")
repetition_penalty = gr.Slider(minimum=0, maximum=2, step=0.01, value=1.05, label="Repetition Penalty")
max_tokens = gr.Slider(minimum=1, maximum=4096, step=1, value=512, label="Max Tokens")
submit = gr.Button("▶")
maintain = gr.Button("☁️")
with gr.Column():
output = gr.Textbox(lines=1, value="", label="Output")
submit.click(fn=generate, inputs=[input, instruction, sampling, temperature, top_p, top_k, repetition_penalty, max_tokens], outputs=[output], queue=False)
maintain.click(cloud, inputs=[], outputs=[], queue=False)
main.launch(show_api=True) |