如何解决自定义函数计算平均绝对偏差
我有一个类似于以下内容的4D numpy数组：
>>>import numpy as np
>>>from functools import partial
>>>X = np.random.rand(20,1,10,4)
>>>X.shape
(20,4)
我计算以下统计信息mean,median,std,p25,p75
>>>percentiles = tuple(partial(np.percentile,q=q) for q in (25,75))
>>>stat_functions = (np.mean,np.std,np.median) + percentiles
>>>stats = np.concatenate([func(X,axis=2,keepdims=True) for func in stat_functions],axis=2)
因此：
>>>stats.shape
(20,5,4)
>>>stats[0]
array([[[0.55187202,0.55892688,0.45816177,0.6378181 ],[0.31028278,0.32109677,0.17319351,0.13341651],[0.57112019,0.60587194,0.45490572,0.59787335],[0.30857011,0.30367621,0.28899686,0.55742753],[0.80678815,0.82014851,0.61295181,0.70529412]]])
我对统计数据中的mad感兴趣，因此我定义了此函数，因为numpy无法使用此函数。
def mad(data):
mean = np.mean(data)
f = lambda x: abs(x - mean)
vf = np.vectorize(f)
return (np.add.reduce(vf(data))) / len(data)
但是我在使此功能起作用时遇到了问题：首先，我尝试过：
>>>stat_functions = (np.mean,np.median,mad) + percentiles
>>>stats = np.concatenate([func(X,axis=2)
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
<ipython-input-33-fa6d972f0fce> in <module>()
----> 1 stats = np.concatenate([func(X,axis=2)
<ipython-input-33-fa6d972f0fce> in <listcomp>(.0)
----> 1 stats = np.concatenate([func(X,axis=2)
TypeError: mad() got an unexpected keyword argument 'axis'
然后我将mad的定义修改为：
def mad(data,axis=None):
...
关注此问题：
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
<ipython-input-35-c74d9e3d057b> in <module>()
----> 1 stats = np.concatenate([f(X,keepdims=True) for f in my_func],axis=2)
<ipython-input-35-c74d9e3d057b> in <listcomp>(.0)
----> 1 stats = np.concatenate([f(X,axis=2)
TypeError: mad() got an unexpected keyword argument 'keepdims'
也这样做：
def mad(data,axis=None,keepdims=None):
...
让我进入：
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-6-c74d9e3d057b> in <module>()
----> 1 stats = np.concatenate([f(X,axis=2)
<__array_function__ internals> in concatenate(*args,**kwargs)
ValueError: all the input arrays must have same number of dimensions,but the array at index 0 has 4 dimension(s) and the array at index 3 has 3 dimension(s)
我知道这与尺寸问题有关，但是我不确定在这种情况下如何解决它。
* 编辑：
根据给出的答案，使用答案的mad函数后，我得到了一个奇怪的结果，如下所示：
stat_functions = (np.mean,mad) + percentiles
stats = np.concatenate([func(X,axis=2)
stats.shape
(20,15,4)
预期输出应为(20,6,4)形状，因为我要在第三个维度中向其中添加一个统计值：(np.mean,mad) + percentiles
EDIT-2
从答案中使用此功能：
def mad(data,axis=-1,keepdims=True):
return np.abs(data - data.mean(axis,keepdims=True)).mean(axis)
然后：
stat_functions = (np.mean,axis=2)
然后碰到这个：
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-16-fa6d972f0fce> in <module>()
----> 1 stats = np.concatenate([func(X,but the array at index 0 has 4 dimension(s) and the array at index 3 has 3 dimension(s)
解决方法
我在代码{
"timestamp": "2020-09-21T15:34:41.367+00:00","status": 400,"error": "Bad Request","trace": "org.springframework.http.converter.HttpMessageNotReadableException: JSON parse error: Cannot deserialize instance of `[Ljava.lang.String;` out of START_OBJECT token; nested exception is com.fasterxml.jackson.databind.exc.MismatchedInputException: Cannot deserialize instance of `[Ljava.lang.String;` out of START_OBJECT token\n at [Source: (PushbackInputStream); line: 1,column: 1]\r\n\tat org.springframework.http.converter.json.AbstractJackson2HttpMessageConverter.readJavaType(AbstractJackson2HttpMessageConverter.java:275)\r\n\tat org.springframework.http.converter.json.AbstractJackson2HttpMessageConverter.read(AbstractJackson2HttpMessageConverter.java:257)\r\n\tat org.springframework.web.servlet.mvc.method.annotation.AbstractMessageConverterMethodArgumentResolver.readWithMessageConverters(AbstractMessageConverterMethodArgumentResolver.java:205)\r\n\tat org.springframework.web.servlet.mvc.method.annotation.RequestResponseBodyMethodProcessor.readWithMessageConverters(RequestResponseBodyMethodProcessor.java:158)\r\n\tat org.springframework.web.servlet.mvc.method.annotation.RequestResponseBodyMethodProcessor.resolveArgument(RequestResponseBodyMethodProcessor.java:131)\r\n\tat org.springframework.web.method.support.HandlerMethodArgumentResolverComposite.resolveArgument(HandlerMethodArgumentResolverComposite.java:121)\r\n\tat org.springframework.web.method.support.InvocableHandlerMethod.getMethodArgumentValues(InvocableHandlerMethod.java:167)\r\n\tat org.springframework.web.method.support.InvocableHandlerMethod.invokeForRequest(InvocableHandlerMethod.java:134)\r\n\tat org.springframework.web.servlet.mvc.method.annotation.ServletInvocableHandlerMethod.invokeAndHandle(ServletInvocableHandlerMethod.java:105)\r\n\tat org.springframework.web.servlet.mvc.method.annotation.RequestMappingHandlerAdapter.invokeHandlerMethod(RequestMappingHandlerAdapter.java:879)\r\n\tat org.springframework.web.servlet.mvc.method.annotation.RequestMappingHandlerAdapter.handleInternal(RequestMappingHandlerAdapter.java:793)\r\n\tat org.springframework.web.servlet.mvc.method.AbstractHandlerMethodAdapter.handle(AbstractHandlerMethodAdapter.java:87)\r\n\tat org.springframework.web.servlet.DispatcherServlet.doDispatch(DispatcherServlet.java:1040)\r\n\tat org.springframework.web.servlet.DispatcherServlet.doService(DispatcherServlet.java:943)\r\n\tat org.springframework.web.servlet.FrameworkServlet.processRequest(FrameworkServlet.java:1006)\r\n\tat org.springframework.web.servlet.FrameworkServlet.doPost(FrameworkServlet.java:909)\r\n\tat javax.servlet.http.HttpServlet.service(HttpServlet.java:660)\r\n\tat org.springframework.web.servlet.FrameworkServlet.service(FrameworkServlet.java:883)\r\n\tat javax.servlet.http.HttpServlet.service(HttpServlet.java:741)\r\n\tat org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:231)\r\n\tat org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:166)\r\n\tat org.apache.tomcat.websocket.server.WsFilter.doFilter(WsFilter.java:53)\r\n\tat org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:193)\r\n\tat org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:166)\r\n\tat org.springframework.web.filter.RequestContextFilter.doFilterInternal(RequestContextFilter.java:100)\r\n\tat org.springframework.web.filter.OncePerRequestFilter.doFilter(OncePerRequestFilter.java:119)\r\n\tat org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:193)\r\n\tat org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:166)\r\n\tat org.spri
中注意到的第一件事绝不是向量化功能（请参见Numpy's doc中的注释。您可以使用vf代替np.abs，您的函数将被矢量化。
也就是说，您的函数可以写为：
abs现在，请注意，此def mad(data):
return np.abs(data - data.mean(0))/ len(data)
函数或您的原始函数仅接受一个位置参数和否可选参数。您收到的错误是因为您试图将mad传递给axis=2：
mad要解决此问题，请使用可选参数构建函数：
[func(X,axis=2,keepdims=True) for func in stat_functions]
或者使用def mad(data,axis=-1,keepdims=True):
return np.abs(data - data.mean(axis,keepdims=keepdims)).sum(axis)/len(data)
比使用mean(axis)
sum(axis)/len(data)版权声明：本文内容由互联网用户自发贡献，该文观点与技术仅代表作者本人。本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如发现本站有涉嫌侵权/违法违规的内容， 请发送邮件至 dio@foxmail.com 举报，一经查实，本站将立刻删除。
