Duplicate from lykeven/visualglm-6b

Co-authored-by: Yan <[email protected]>

.gitattributes

@@ -0,0 +1,34 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,13 @@
+---
+title: Visualglm-6b
+emoji: 💻
+colorFrom: purple
+colorTo: pink
+sdk: gradio
+sdk_version: 3.27.0
+app_file: app.py
+pinned: false
+duplicated_from: lykeven/visualglm-6b
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,209 @@
+#!/usr/bin/env python
+
+import gradio as gr
+import os
+import re
+from PIL import Image
+import base64
+import time
+
+DESCRIPTION = '''# <a href="https://github.com/THUDM/VisualGLM-6B">VisualGLM</a>'''
+
+MAINTENANCE_NOTICE1 = 'Hint 1: If the app report "Something went wrong, connection error out", please turn off your proxy and retry.\nHint 2: If you upload a large size of image like 10MB, it may take some time to upload and process. Please be patient and wait.'
+MAINTENANCE_NOTICE2 = '提示1: 如果应用报了“Something went wrong, connection error out”的错误，请关闭代理并重试。\n提示2: 如果你上传了很大的图片，比如10MB大小，那将需要一些时间来上传和处理，请耐心等待。'
+
+NOTES = 'This app is adapted from <a href="https://github.com/THUDM/VisualGLM-6B">https://github.com/THUDM/VisualGLM-6B</a>. It would be recommended to check out the repo if you want to see the detail of our model.'
+
+import json
+import requests
+import base64
+import hashlib
+
+default_chatbox = [("", "Hi, What do you want to know about this image?")]
+
+URL = os.environ.get("URL")
+
+def process_image(image_prompt):
+    image = Image.open(image_prompt)
+    print(f"height:{image.height}, width:{image.width}")
+    resized_image = image.resize((224, 224), )
+    timestamp = int(time.time())
+    file_ext = os.path.splitext(image_prompt)[1]
+    filename = f"examples/{timestamp}{file_ext}"
+    resized_image.save(filename)
+    print(f"temporal filename {filename}")
+    with open(filename, "rb") as image_file:
+        bytes = base64.b64encode(image_file.read())
+        encoded_img = str(bytes, encoding='utf-8')
+        image_hash = hashlib.sha256(bytes).hexdigest()
+        os.remove(filename)
+        return encoded_img, image_hash
+
+
+def process_image_without_resize(image_prompt):
+    image = Image.open(image_prompt)
+    print(f"height:{image.height}, width:{image.width}")
+    timestamp = int(time.time())
+    file_ext = os.path.splitext(image_prompt)[1]
+    filename = f"examples/{timestamp}{file_ext}"
+    image.save(filename)
+    print(f"temporal filename {filename}")
+    with open(filename, "rb") as image_file:
+        bytes = base64.b64encode(image_file.read())
+        encoded_img = str(bytes, encoding='utf-8')
+        image_hash = hashlib.sha256(bytes).hexdigest()
+        os.remove(filename)
+        return encoded_img, image_hash
+
+
+def is_chinese(text):
+    zh_pattern = re.compile(u'[\u4e00-\u9fa5]+')
+    return zh_pattern.search(text)
+
+
+def post(
+        input_text,
+        temperature,
+        top_p,
+        image_prompt,
+        result_previous,
+        hidden_image        
+        ):
+    result_text = [(ele[0], ele[1]) for ele in result_previous]
+    for i in range(len(result_text)-1, -1, -1):
+        if result_text[i][0] == "":
+            del result_text[i]
+    print(f"history {result_text}")
+
+    is_zh = is_chinese(input_text)
+
+    if image_prompt is None:
+        print("Image empty")
+        if is_zh:
+            result_text.append((input_text, '图片为空！请上传图片并重试。'))
+        else:
+            result_text.append((input_text, 'Image empty! Please upload a image and retry.'))
+        return input_text, result_text, hidden_image
+    elif input_text == "":
+        print("Text empty")
+        result_text.append((input_text, 'Text empty! Please enter text and retry.'))
+        return "", result_text, hidden_image                
+
+    headers = {
+            "Content-Type": "application/json; charset=UTF-8",
+            "User-Agent": "Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/67.0.3396.87 Safari/537.36",
+        }
+    if image_prompt:
+        encoded_img, image_hash = process_image_without_resize(image_prompt)
+        print(f"image_hash:{image_hash}, hidden_image_hash:{hidden_image}")
+
+        if hidden_image is not None and image_hash != hidden_image:
+            print("image has been update")
+            result_text = []        
+        hidden_image = image_hash        
+    else:
+        encoded_img = None
+
+    print('开始请求...')
+    data = json.dumps({
+        'text': input_text,
+        'image_prompt': encoded_img,
+        'temperature': temperature,
+        'top_p': top_p,
+        'history': result_text
+    })
+    try:
+        response = requests.request("POST", URL, headers=headers, data=data, timeout=(60, 100)).json()
+    except Exception as e:
+        print("error message", e)
+        if is_zh:
+            result_text.append((input_text, '超时！请稍等几分钟再重试。'))
+        else:
+            result_text.append((input_text, 'Timeout! Please wait a few minutes and retry.'))
+        return "", result_text, hidden_image
+    print('请求完毕...')
+    # response = {'result':input_text}
+
+    answer = str(response['result'])
+    result_text.append((input_text, answer))
+    print(result_text)
+    print('finished')
+    return "", result_text, hidden_image
+
+
+def clear_fn(value):
+    return "", default_chatbox, None
+
+def clear_fn2(value):
+    return default_chatbox
+
+def io_fn(a, b, c):
+    print(f"call io_fn")
+    return a, b
+
+
+def change_language(value):
+    if value == "Change hint to English":
+        return "提示变为中文", MAINTENANCE_NOTICE1
+    else:
+        return "Change hint to English", MAINTENANCE_NOTICE2
+
+
+def main():
+    gr.close_all()
+    examples = []
+    with open("./examples/example_inputs.jsonl") as f:
+        for line in f:
+            data = json.loads(line)
+            examples.append(data)
+
+
+    with gr.Blocks(css='style.css') as demo:
+
+        with gr.Row():
+            with gr.Column(scale=4.5):
+                with gr.Group():
+                    input_text = gr.Textbox(label='Input Text', placeholder='Please enter text prompt below and press ENTER.')
+                    with gr.Row():
+                        run_button = gr.Button('Generate')
+                        clear_button = gr.Button('Clear')
+
+                    image_prompt = gr.Image(type="filepath", label="Image Prompt", value=None)
+                with gr.Row():
+                    temperature = gr.Slider(maximum=1, value=0.8, minimum=0, label='Temperature')
+                    top_p = gr.Slider(maximum=1, value=0.4, minimum=0, label='Top P')
+                with gr.Group():
+                    with gr.Row():
+                        with gr.Column(scale=7):    
+                            maintenance_notice = gr.Markdown(MAINTENANCE_NOTICE1)
+                        with gr.Column(scale=2):    
+                            change_button = gr.Button('Change hint to English', visible=False)
+            with gr.Column(scale=5.5):
+                result_text = gr.components.Chatbot(label='Multi-round conversation History', value=[("", "Hi, What do you want to know about this image?")]).style(height=550)
+                hidden_image_hash = gr.Textbox(visible=False)
+
+        gr_examples = gr.Examples(examples=[[example["text"], example["image"]] for example in examples], 
+                                  inputs=[input_text, image_prompt],
+                                  label="Example Inputs (Click to insert an examplet into the input box)",
+                                  examples_per_page=3)
+
+        gr.Markdown(NOTES)
+
+        print(gr.__version__)
+        run_button.click(fn=post,inputs=[input_text, temperature, top_p, image_prompt, result_text, hidden_image_hash],
+                         outputs=[input_text, result_text, hidden_image_hash])
+        input_text.submit(fn=post,inputs=[input_text, temperature, top_p, image_prompt, result_text, hidden_image_hash],
+                         outputs=[input_text, result_text, hidden_image_hash])
+        clear_button.click(fn=clear_fn, inputs=clear_button, outputs=[input_text, result_text, image_prompt])
+        image_prompt.upload(fn=clear_fn2, inputs=clear_button, outputs=[result_text])
+        image_prompt.clear(fn=clear_fn2, inputs=clear_button, outputs=[result_text])
+
+        print(gr.__version__)
+
+
+    demo.queue(concurrency_count=10)
+    demo.launch()
+
+
+if __name__ == '__main__':
+    main()

examples/example_inputs.jsonl

@@ -0,0 +1,3 @@
+{"id":1, "text": "描述一下这个场景", "image": "examples/1.jpeg"}
+{"id":2, "text": "What object it is?", "image": "examples/2.jpeg"}
+{"id":3, "text": "这张图片描述了什么", "image": "examples/3.jpeg"}

style.css

@@ -0,0 +1,7 @@
+h1 {
+  text-align: center;
+}
+img#visitor-badge {
+  display: block;
+  margin: auto;
+}