使用Vectorbt绘制RSI时DOMWidget属性错误的解决求助

问题场景 #

在Python 3.10.9（Vectorbt支持的最新版本）环境下，使用Vectorbt库在Jupyter Notebook中绘制简单RSI指标，图表可正常生成，但运行代码时触发AttributeError: type object 'DOMWidget' has no attribute '_ipython_display_'错误。已尝试升级Plotly至最新版本，问题仍未解决。

相关代码 #

#nilagyan ng r para raw file dahil nababasa ng python as UNICODE pag may slash tapos U ang kasunod
df = pd.read_csv(r"C:\Users\Mark\Documents\Trading\BTC-2021min\BTC-2021min.csv")

#dahil ang unix code ay about sa time, cinonvert sa seconds para magkaroon ng column na date
#basically nag add tayo ng COLUMN or Data Frame na pangalan ay date
df["date"] = pd.to_datetime(df["unix"], unit = "s")

#nanggawa ng column at index na date at time kaya nakaSET ang INDEX sa DATE
df.set_index("date", inplace=True)

#aggregation ng 1 minute candles into something bigger, sakin ay 15 minutes kaya 15T
df = df.resample("15T").agg({
    "open":"first",
    "high": "max",
    "low": "min",
    "close": "last",
})

#this makes it go through only a specific date rather than the whole csv file, start date at end date lang 
df = df.loc["2021-10-01":"2021-10-31"]


rsi = vbt.RSI.run(df.close, window=8)

rsi.plot()

报错堆栈 #

AttributeError                            Traceback (most recent call last)
File c:\Users\Mark\AppData\Local\Programs\Python\Python310\lib\site-packages\IPython\core\formatters.py:920, in IPythonDisplayFormatter.__call__(self, obj)
    918 method = get_real_method(obj, self.print_method)
    919 if method is not None:
---> 920     method()
    921     return True

File c:\Users\Mark\AppData\Local\Programs\Python\Python310\lib\site-packages\plotly\basewidget.py:741, in BaseFigureWidget._ipython_display_(self)
    737 """
    738 Handle rich display of figures in ipython contexts
    739 """
    740 # Override BaseFigure's display to make sure we display the widget version
---> 741 widgets.DOMWidget._ipython_display_(self)

AttributeError: type object 'DOMWidget' has no attribute '_ipython_display_'

解决方法 #

• 降级ipywidgets到兼容版本
  该错误源于ipywidgets 8.x+版本对DOMWidget的方法做了重构，而Plotly的basewidget仍依赖旧方法。执行以下命令降级：

  pip install ipywidgets==7.7.1
  

  安装完成后重启Jupyter Notebook即可。

• 强制使用静态渲染模式
  绕开Widget渲染逻辑，直接使用Plotly的静态渲染器。在代码开头添加配置：

  import plotly.io as pio
  # 根据你使用的环境选择渲染器，Notebook用'notebook'，JupyterLab用'jupyterlab'
  pio.renderers.default = 'notebook'
  

  或者直接在plot方法中指定渲染器：

  rsi.plot(renderer='notebook')
  

• 同步Vectorbt与Plotly版本
  查看Vectorbt官方要求的Plotly版本范围，执行pip show vectorbt查看Requires字段，安装符合要求的Plotly版本，避免版本过高导致不兼容。例如：

  # 假设Vectorbt要求Plotly>=5.0,<5.15
  pip install plotly==5.14.1
  

内容的提问来源于stack exchange，提问作者zcodec