upload

app.py

@@ -0,0 +1,249 @@
+# Copyright (c) Alibaba, Inc. and its affiliates.
+import os
+
+import gradio as gr
+import modelscope_studio.components.antd as antd
+import modelscope_studio.components.base as ms
+from PIL import Image
+import secrets
+import tempfile
+from http import HTTPStatus
+from urllib3.exceptions import HTTPError
+
+from pathlib import Path
+
+os.environ['DASHSCOPE_HTTP_BASE_URL'] = 'https://dashscope.aliyuncs.com/api/v1'
+# os.environ['DASHSCOPE_WEBSOCKET_BASE_URL'] = 'https://poc-dashscope.aliyuncs.com/api-ws/v1/inference'
+
+import dashscope
+from dashscope import MultiModalConversation
+API_KEY = os.environ['API_KEY']
+dashscope.api_key = API_KEY
+
+is_modelscope_studio = os.getenv('MODELSCOPE_ENVIRONMENT') == 'studio'
+
+def get_text(text: str, cn_text: str):
+    if is_modelscope_studio:
+        return cn_text
+    return text
+
+def resolve_image(filename):
+    return os.path.join(os.path.dirname(__file__), filename)
+
+DEMO_LIST = [
+  {
+    "description": "Evaluate the integral of the functions graphed using the formula for circles: ",
+    "image": resolve_image("./examples/1.webp")
+  },
+  {
+    "description": "回答图中问题",
+    "image": resolve_image("./examples/2.png")
+  },
+  {
+    "description": "图片中的滤液E是什么化学物质?",
+    "image": resolve_image("./examples/3.png")
+  },
+  {
+    "description": "I want to know the volume of this sofa",
+    "image": resolve_image("./examples/4.png")
+  },
+]
+
+def process_image(image, shouldConvert=False):
+    # 获取上传文件的目录
+    uploaded_file_dir = os.environ.get("GRADIO_TEMP_DIR") or str(
+        Path(tempfile.gettempdir()) / "gradio")
+    os.makedirs(uploaded_file_dir, exist_ok=True)
+
+    # 创建临时文件路径
+    name = f"tmp{secrets.token_hex(20)}.jpg"
+    filename = os.path.join(uploaded_file_dir, name)
+    # 保存上传的图片
+    if shouldConvert:
+        new_img = Image.new('RGB',
+                            size=(image.width, image.height),
+                            color=(255, 255, 255))
+        new_img.paste(image, (0, 0), mask=image)
+        image = new_img
+    image.save(filename)
+
+    return filename
+
+
+
+def generate(image, query):
+    imageFile = process_image(image)
+    content = [
+        {'image': f'file://{imageFile}'},
+        {'text': query}
+    ]
+    messages = [
+        {'role': 'user', 'content': content},
+    ]
+    print('messages:', messages)
+    responses = MultiModalConversation.call(
+        model='qvq-72b-preview', messages=messages, stream=True,
+    )
+    for response in responses:
+        if not response.status_code == HTTPStatus.OK:
+            raise HTTPError(f'response.code: {response.code}\nresponse.message: {response.message}')
+        response = response.output.choices[0].message.content
+        if len(response) > 0 and response[0]['text']:
+            print(response[0]['text'])
+            yield response[0]['text']
+
+
+if __name__ == "__main__":
+
+    def on_clear():
+        return {
+            input: gr.update(value=None),
+            **{
+                item: gr.update(value=None)
+                for item in input_image
+            },
+        }
+
+    with gr.Blocks() as demo:
+        with ms.Application() as app:
+            with antd.ConfigProvider(
+                    locale="zh_CN" if is_modelscope_studio else None,
+                    theme=dict(token=dict(colorPrimary="#a855f7"))):
+                with antd.Card(elem_style=dict(marginBottom=12),
+                               styles=dict(body=dict(padding=4))):
+                    with antd.Flex(elem_style=dict(width="100%"),
+                                   justify="center",
+                                   align="center",
+                                   gap=14):
+                        with ms.Div(elem_style=dict(flexShrink=0)):
+                            antd.Image(
+                                resolve_image("./cutelogo.jpg"),
+                                preview=False,
+                                height=60)
+                        with ms.Div():
+                            antd.Typography.Title(
+                                "QVQ-72B-Preview",
+                                elem_style=dict(margin=0, fontSize=24),
+                                level=1)
+                with ms.AutoLoading():
+                    with antd.Row(gutter=[8, 8], align="stretch"):
+                        with antd.Col(xs=24, md=8):
+                            with antd.Space(direction="vertical",
+                                            elem_style=dict(width="100%")):
+                                with antd.Space(direction="vertical",
+                                                elem_style=dict(width="100%"),
+                                                elem_id="input-container"):
+                                    with ms.Fragment():
+                                        input_image = gr.Image(
+                                                    type="pil",
+                                                    label="Upload",
+                                                    sources=["upload"]),
+                                    input = antd.Input.Textarea(
+                                        placeholder=get_text("Ask a question", "输入一个问题"),
+                                        auto_size=dict(maxRows=6, minRows=2),
+                                        allow_clear=True)
+
+                                with antd.Flex(align="center",
+                                               justify="space-between"):
+                                    antd.Typography.Text(
+                                        get_text("Warning: This model only supports single-turn dialogue.",  "注：当前模型只支持单轮对话，如需中文回答，提示词加“用中文回答”"), type="warning")
+                                    tour_btn = antd.Button(get_text("Tour", "使用指引"),
+                                                           variant="filled",
+                                                           color="default")
+
+                                with antd.Row(gutter=8):
+                                    with antd.Col(span=12):
+                                        clear_btn = antd.Button(get_text("Clear", "清除"),
+                                                                block=True)
+                                    with antd.Col(span=12):
+                                        submit_btn = antd.Button(
+                                            get_text("Submit", "提交"),
+                                            type="primary",
+                                            block=True,
+                                            elem_id="submit-btn")
+
+                                antd.Divider(get_text("Example", "示例"))
+
+                                with antd.Flex(gap="small", wrap=True):
+                                    for item in DEMO_LIST:
+
+                                        def bind_on_example(_item):
+                                            def on_example():
+                                                return gr.update(
+                                                        value=_item[
+                                                            'description']
+                                                    ), gr.update(
+                                                        value=_item['image'])
+
+                                            return on_example
+
+                                        with antd.Card(
+                                                hoverable=True,
+                                                elem_style=dict(
+                                                    width="100%")) as example:
+                                            if "description" in item:
+                                                antd.Typography.Text(
+                                                    item["description"])
+                                            if "image" in item:
+                                                antd.Image(item["image"],
+                                                           preview=False)
+                                        example.click(
+                                            fn=bind_on_example(item),
+                                            outputs=[input, input_image[0]])
+
+                        with antd.Col(xs=24, md=16):
+                            with antd.Card(title=get_text("Answer", "答案"),
+                                           elem_style=dict(height="100%"),
+                                           elem_id="output-container"):
+                                output = gr.Markdown(
+                                    show_copy_button=True,
+                                    latex_delimiters=[{
+                                        "left": '$$',
+                                        "right": '$$',
+                                        "display": True
+                                    }, {
+                                        "left": '$',
+                                        "right": '$',
+                                        "display": False,
+                                    }, {
+                                        "left": '\\(',
+                                        "right": '\\)',
+                                        "display": False,
+                                    }, {
+                                        "left": '\\[',
+                                        "right": '\\]',
+                                        "display": True
+                                    }])
+                    with antd.Tour(props=dict(open=False)) as tour:
+                        antd.Tour.Step(
+                            title=get_text("Step 1", "步骤 1"),
+                            description=get_text("Upload image and enter text", "传入图片和文本"),
+                            get_target=
+                            "() => document.querySelector('#input-container')")
+                        antd.Tour.Step(
+                            title=get_text("Step 2","步骤 2"),
+                            description=get_text("Click submit button", "点击提交按钮"),
+                            get_target=
+                            "() => document.querySelector('#submit-btn')")
+                        antd.Tour.Step(
+                            title=get_text("Step 3","步骤 3"),
+                            description=get_text("Wait for result", "等待结果返回"),
+                            get_target=
+                            "() => document.querySelector('#output-container')"
+                        )
+
+                    tour_btn.click(fn=lambda: gr.update(props=dict(open=True)),
+                                   outputs=[tour])
+                    gr.on([tour.finish, tour.close],
+                          fn=lambda: gr.update(props=dict(open=False)),
+                          outputs=[tour])
+
+                    submit_btn.click(
+                        fn=generate,
+                        inputs=[*input_image, input],
+                        outputs=[output])
+                    clear_btn.click(
+                        fn=on_clear,
+                        outputs=[*input_image, input])
+
+                demo.queue(default_concurrency_limit=50).launch()

app_bak.py

@@ -0,0 +1,236 @@
+# Copyright (c) Alibaba Cloud.
+#
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+import os
+import numpy as np
+from urllib3.exceptions import HTTPError
+os.environ['DASHSCOPE_HTTP_BASE_URL'] = 'https://poc-dashscope.aliyuncs.com/api/v1'
+os.environ['DASHSCOPE_WEBSOCKET_BASE_URL'] = 'https://poc-dashscope.aliyuncs.com/api-ws/v1/inference'
+
+from argparse import ArgumentParser
+from pathlib import Path
+
+import copy
+import gradio as gr
+import oss2
+import os
+import re
+import secrets
+import tempfile
+import requests
+from http import HTTPStatus
+from dashscope import MultiModalConversation
+import dashscope
+API_KEY = os.environ['API_KEY']
+dashscope.api_key = API_KEY
+
+REVISION = 'v1.0.4'
+BOX_TAG_PATTERN = r"<box>([\s\S]*?)</box>"
+PUNCTUATION = "！？。＂＃＄％＆＇（）＊＋，－／：；＜＝＞＠［＼］＾＿｀｛｜｝～｟｠｢｣､、〃》「」『』【】〔〕〖〗〘〙〚〛〜〝〞〟〰〾〿–—‘’‛“”„‟…‧﹏."
+
+
+def _get_args():
+    parser = ArgumentParser()
+    parser.add_argument("--revision", type=str, default=REVISION)
+    parser.add_argument("--cpu-only", action="store_true", help="Run demo with CPU only")
+
+    parser.add_argument("--share", action="store_true", default=False,
+                        help="Create a publicly shareable link for the interface.")
+    parser.add_argument("--inbrowser", action="store_true", default=False,
+                        help="Automatically launch the interface in a new tab on the default browser.")
+    parser.add_argument("--server-port", type=int, default=7860,
+                        help="Demo server port.")
+    parser.add_argument("--server-name", type=str, default="127.0.0.1",
+                        help="Demo server name.")
+
+    args = parser.parse_args()
+    return args
+
+def _parse_text(text):
+    lines = text.split("\n")
+    lines = [line for line in lines if line != ""]
+    count = 0
+    for i, line in enumerate(lines):
+        if "```" in line:
+            count += 1
+            items = line.split("`")
+            if count % 2 == 1:
+                lines[i] = f'<pre><code class="language-{items[-1]}">'
+            else:
+                lines[i] = f"<br></code></pre>"
+        else:
+            if i > 0:
+                if count % 2 == 1:
+                    line = line.replace("`", r"\`")
+                    line = line.replace("<", "&lt;")
+                    line = line.replace(">", "&gt;")
+                    line = line.replace(" ", "&nbsp;")
+                    line = line.replace("*", "&ast;")
+                    line = line.replace("_", "&lowbar;")
+                    line = line.replace("-", "&#45;")
+                    line = line.replace(".", "&#46;")
+                    line = line.replace("!", "&#33;")
+                    line = line.replace("(", "&#40;")
+                    line = line.replace(")", "&#41;")
+                    line = line.replace("$", "&#36;")
+                lines[i] = "<br>" + line
+    text = "".join(lines)
+    return text
+
+
+def _remove_image_special(text):
+    text = text.replace('<ref>', '').replace('</ref>', '')
+    return re.sub(r'<box>.*?(</box>|$)', '', text)
+
+def _launch_demo(args):
+    uploaded_file_dir = os.environ.get("GRADIO_TEMP_DIR") or str(
+        Path(tempfile.gettempdir()) / "gradio"
+    )
+
+    def predict(_chatbot, task_history):
+        chat_query = _chatbot[-1][0]
+        query = task_history[-1][0]
+        if len(chat_query) == 0:
+            _chatbot.pop()
+            task_history.pop()
+            return _chatbot
+        print("User: " + _parse_text(query))
+        history_cp = copy.deepcopy(task_history)
+        full_response = ""
+        messages = []
+        content = []
+        for q, a in history_cp:
+            if isinstance(q, (tuple, list)):
+                content.append({'image': f'file://{q[0]}'})
+            else:
+                content.append({'text': q})
+                messages.append({'role': 'user', 'content': content})
+                messages.append({'role': 'assistant', 'content': [{'text': a}]})
+                content = []
+        messages.pop()
+        responses = MultiModalConversation.call(
+            model='pre-qvq-72b-preview-1219', messages=messages, stream=True,
+        )
+        for response in responses:
+            if not response.status_code == HTTPStatus.OK:
+                raise HTTPError(f'response.code: {response.code}\nresponse.message: {response.message}')
+            response = response.output.choices[0].message.content
+            response_text = []
+            for ele in response:
+                if 'text' in ele:
+                    response_text.append(ele['text'])
+                elif 'box' in ele:
+                    response_text.append(ele['box'])
+            response_text = ''.join(response_text)
+            _chatbot[-1] = (_parse_text(chat_query), _remove_image_special(response_text))
+            yield _chatbot
+
+        if len(response) > 1:
+            result_image = response[-1]['result_image']
+            resp = requests.get(result_image)
+            os.makedirs(uploaded_file_dir, exist_ok=True)
+            name = f"tmp{secrets.token_hex(20)}.jpg"
+            filename = os.path.join(uploaded_file_dir, name)
+            with open(filename, 'wb') as f:
+                f.write(resp.content)
+            response = ''.join(r['box'] if 'box' in r else r['text'] for r in response[:-1])
+            _chatbot.append((None, (filename,)))
+        else:
+            response = response[0]['text']
+            _chatbot[-1] = (_parse_text(chat_query), response)
+        full_response = _parse_text(response)
+
+        task_history[-1] = (query, full_response)
+        print("Qwen2-VL-Chat: " + _parse_text(full_response))
+        yield _chatbot
+
+
+    def regenerate(_chatbot, task_history):
+        if not task_history:
+            return _chatbot
+        item = task_history[-1]
+        if item[1] is None:
+            return _chatbot
+        task_history[-1] = (item[0], None)
+        chatbot_item = _chatbot.pop(-1)
+        if chatbot_item[0] is None:
+            _chatbot[-1] = (_chatbot[-1][0], None)
+        else:
+            _chatbot.append((chatbot_item[0], None))
+        _chatbot_gen = predict(_chatbot, task_history)
+        for _chatbot in _chatbot_gen:
+            yield _chatbot
+
+    def add_text(history, task_history, text):
+        task_text = text
+        history = history if history is not None else []
+        task_history = task_history if task_history is not None else []
+        history = history + [(_parse_text(text), None)]
+        task_history = task_history + [(task_text, None)]
+        return history, task_history, ""
+
+    def add_file(history, task_history, file):
+        history = history if history is not None else []
+        task_history = task_history if task_history is not None else []
+        history = history + [((file.name,), None)]
+        task_history = task_history + [((file.name,), None)]
+        return history, task_history
+
+    def reset_user_input():
+        return gr.update(value="")
+
+    def reset_state(task_history):
+        task_history.clear()
+        return []
+
+    with gr.Blocks() as demo:
+        gr.Markdown("""\
+<p align="center"><img src="https://modelscope.oss-cn-beijing.aliyuncs.com/resource/qwen.png" style="height: 80px"/><p>""")
+        gr.Markdown("""<center><font size=8>QVQ-72B-preview</center>""")
+        gr.Markdown(
+            """\
+<center><font size=3>This WebUI is based on QVQ-72B-preview, developed by Alibaba Cloud.</center>""")
+        gr.Markdown("""<center><font size=3>本WebUI基于Qwen2-VL-Max。</center>""")
+
+        chatbot = gr.Chatbot(label='QVQ-72B-preview', elem_classes="control-height", height=500)
+        query = gr.Textbox(label='Input', )
+        task_history = gr.State([])
+
+        with gr.Row():
+            addfile_btn = gr.UploadButton("📁 Upload (上传文件)", file_types=["image"])
+            submit_btn = gr.Button("🚀 Submit (发送)")
+            regen_btn = gr.Button("🤔️ Regenerate (重试)")
+
+        query.submit(add_text, [chatbot, task_history, query], [chatbot, task_history]).then(
+            predict, [chatbot, task_history], [chatbot], show_progress=True
+        )
+        submit_btn.click(add_text, [chatbot, task_history, query], [chatbot, task_history]).then(
+            predict, [chatbot, task_history], [chatbot], show_progress=True
+        )
+        submit_btn.click(reset_user_input, [], [query])
+        regen_btn.click(regenerate, [chatbot, task_history], [chatbot], show_progress=True)
+        addfile_btn.upload(add_file, [chatbot, task_history, addfile_btn], [chatbot, task_history], show_progress=True)
+
+        gr.Markdown("""\
+<font size=2>Note: This demo is governed by the original license of Qwen2-VL. \
+We strongly advise users not to knowingly generate or allow others to knowingly generate harmful content, \
+including hate speech, violence, pornography, deception, etc. \
+(注：本演示受Qwen2-VL的许可协议限制。我们强烈建议，用户不应传播及不应允许他人传播以下内容，\
+包括但不限于仇恨言论、暴力、色情、欺诈相关的有害信息。)""")
+
+    demo.queue().launch(
+        share=args.share,
+        # inbrowser=args.inbrowser,
+        # server_port=args.server_port,
+        # server_name=args.server_name,
+    )
+
+
+def main():
+    args = _get_args()
+    _launch_demo(args)
+
+
+if __name__ == '__main__':
+    main()

requirements.txt

@@ -0,0 +1,3 @@
+dashscope
+oss2
+modelscope_studio