Python 将sparaFrame保存到CSV会引发索引器错误 用例子更新了问题

我试着重复我的问题。事实证明，它甚至与我的数据集的大小无关。这是重现我的问题的一个最小示例：

>>> import pandas as pd
>>> data = pd.SparseDataFrame({ 'user': ['a', 'b', 'c', 'd'], 'week': [4, 3, 2, 1] }, default_fill_value=0)
>>> data.info()
<class 'pandas.sparse.frame.SparseDataFrame'>
RangeIndex: 4 entries, 0 to 3
Data columns (total 2 columns):
user    4 non-null object
week    4 non-null int64
dtypes: int64(1), object(1)
memory usage: 144.0+ bytes
>>> data.to_csv('error.csv', index=False)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/usr/local/lib/python3.6/site-packages/pandas/core/frame.py", line 1383, in to_csv
    formatter.save()
  File "/usr/local/lib/python3.6/site-packages/pandas/formats/format.py", line 1475, in save
    self._save()
  File "/usr/local/lib/python3.6/site-packages/pandas/formats/format.py", line 1576, in _save
    self._save_chunk(start_i, end_i)
  File "/usr/local/lib/python3.6/site-packages/pandas/formats/format.py", line 1590, in _save_chunk
    quoting=self.quoting)
  File "/usr/local/lib/python3.6/site-packages/pandas/core/internals.py", line 596, in to_native_types
    values = values[:, slicer]
  File "/usr/local/lib/python3.6/site-packages/pandas/sparse/array.py", line 401, in __getitem__
    data_slice = self.values[key]
IndexError: too many indices for array

当我试图将其保存到CSV文件时，它会引发一个

索引器

。这是因为数据太大了吗？指定

chunksize

无法解决此问题

>>> data.to_csv('../data/hashtags_binarized.csv', index=False)

---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
<ipython-input-58-550cc98888dc> in <module>()
----> 1 get_ipython().run_cell_magic('time', '', "data.to_csv('../data/hashtags_binarized.csv', index=False)")

/usr/local/lib/python3.6/site-packages/IPython/core/interactiveshell.py in run_cell_magic(self, magic_name, line, cell)
   2113             magic_arg_s = self.var_expand(line, stack_depth)
   2114             with self.builtin_trap:
-> 2115                 result = fn(magic_arg_s, cell)
   2116             return result
   2117 

<decorator-gen-59> in time(self, line, cell, local_ns)

/usr/local/lib/python3.6/site-packages/IPython/core/magic.py in <lambda>(f, *a, **k)
    186     # but it's overkill for just that one bit of state.
    187     def magic_deco(arg):
--> 188         call = lambda f, *a, **k: f(*a, **k)
    189 
    190         if callable(arg):

/usr/local/lib/python3.6/site-packages/IPython/core/magics/execution.py in time(self, line, cell, local_ns)
   1179         if mode=='eval':
   1180             st = clock2()
-> 1181             out = eval(code, glob, local_ns)
   1182             end = clock2()
   1183         else:

<timed eval> in <module>()

/usr/local/lib/python3.6/site-packages/pandas/core/frame.py in to_csv(self, path_or_buf, sep, na_rep, float_format, columns, header, index, index_label, mode, encoding, compression, quoting, quotechar, line_terminator, chunksize, tupleize_cols, date_format, doublequote, escapechar, decimal)
   1381                                      doublequote=doublequote,
   1382                                      escapechar=escapechar, decimal=decimal)
-> 1383         formatter.save()
   1384 
   1385         if path_or_buf is None:

/usr/local/lib/python3.6/site-packages/pandas/formats/format.py in save(self)
   1473                 self.writer = csv.writer(f, **writer_kwargs)
   1474 
-> 1475             self._save()
   1476 
   1477         finally:

/usr/local/lib/python3.6/site-packages/pandas/formats/format.py in _save(self)
   1574                 break
   1575 
-> 1576             self._save_chunk(start_i, end_i)
   1577 
   1578     def _save_chunk(self, start_i, end_i):

/usr/local/lib/python3.6/site-packages/pandas/formats/format.py in _save_chunk(self, start_i, end_i)
   1588                                   decimal=self.decimal,
   1589                                   date_format=self.date_format,
-> 1590                                   quoting=self.quoting)
   1591 
   1592             for col_loc, col in zip(b.mgr_locs, d):

/usr/local/lib/python3.6/site-packages/pandas/core/internals.py in to_native_types(self, slicer, na_rep, quoting, **kwargs)
    594         values = self.values
    595         if slicer is not None:
--> 596             values = values[:, slicer]
    597         mask = isnull(values)
    598 

/usr/local/lib/python3.6/site-packages/pandas/sparse/array.py in __getitem__(self, key)
    399             return self._get_val_at(key)
    400         elif isinstance(key, tuple):
--> 401             data_slice = self.values[key]
    402         else:
    403             if isinstance(key, SparseArray):

IndexError: too many indices for array

>>data.to_csv（'../data/hashtags_binarized.csv'，index=False）
---------------------------------------------------------------------------
索引器回溯（最后一次最近调用）
在（）
---->1获取\u ipython（）。运行\u cell\u magic（'time'，''，'data.to_csv（'../data/hashtags\u binarized.csv'，index=False'））
/run_cell_magic中的usr/local/lib/python3.6/site-packages/IPython/core/interactiveshell.py（self，magic_name，line，cell）
2113 magic_arg_s=self.var_expand（行、堆栈深度）
2114带自建存水弯：
->2115结果=fn（魔法参数，单元格）
2116返回结果
2117
及时（自身、线路、小区、本地）
/usr/local/lib/python3.6/site-packages/IPython/core/magic.py in（f，*a，**k）
186#但仅仅为了那一点点国家就太过分了。
187 def魔术装饰（arg）：
-->188调用=λf，*a，**k:f（*a，**k）
189
190如果可调用（arg）：
/usr/local/lib/python3.6/site-packages/IPython/core/magics/execution.py及时（self、line、cell、local）
1179如果模式=='eval'：
1180 st=时钟2（）
->1181 out=评估（代码、全局、本地）
1182结束=时钟2（）
1183其他：
在（）
/usr/local/lib/python3.6/site-packages/pandas/core/frame.py in to_csv（self、path或buf、sep、na_rep、float格式、列、标题、索引、索引标签、模式、编码、压缩、引号、行终止符、chunksize、tupleize cols、date格式、双引号、转义、十进制）
1381双引号=双引号，
1382 escapechar=escapechar，十进制=十进制）
->1383格式化程序。保存（）
1384
1385如果路径_或_buf为无：
/保存中的usr/local/lib/python3.6/site-packages/pandas/formats/format.py（self）
1473 self.writer=csv.writer（f，**writer_-kwargs）
1474
->1475自我保存（）
1476
1477最后：
/usr/local/lib/python3.6/site-packages/pandas/formats/format.py in_save（self）
1574休息
1575
->1576自我保存块（开始、结束）
1577
1578 def_save_chunk（self、start_i、end_i）：
/usr/local/lib/python3.6/site-packages/pandas/formats/format.py in_save_chunk（self、start_i、end_i）
1588十进制=自十进制，
1589日期格式=self.date格式，
->1590报价=自报价）
1591
1592对于col_loc，col in zip（b.mgr_locs，d）：
/usr/local/lib/python3.6/site-packages/pandas/core/internals.py到“本地”类型（self、slicer、na_rep、quoting，**kwargs）
594值=自我价值
595如果切片器不是无：
-->596值=值[：，切片器]
597掩码=isnull（值）
598
/usr/local/lib/python3.6/site-packages/pandas/sparse/array.py in_u__获取项目（self，key）
399返回自我。获取价值（键）
400 elif isinstance（键，元组）：
-->401数据片=自身值[键]
402其他：
403如果存在（键，SparseArray）：
索引器：数组的索引太多

使用另一个应用“toCSV”（“name.CSV”）创建CSV的选项，您将得到一个错误“sparStateAFrame”对象没有属性“toCSV”。因此，请使用“.to_dense（）.to_csv（'name.csv'））

---

嘿你能拿出一段数据让我们重新创建错误吗？我添加了一个最小的例子。实际上，

to_pickle

是有效的，所以这可能只是一个bug…很好的发现。是的，我试着看看稀疏数据帧是否有to_csv函数，但找不到任何东西。但这并不意味着你做不到，只是熊猫文档有时候会有点糟糕。

>>> data.to_csv('../data/hashtags_binarized.csv', index=False)

---------------------------------------------------------------------------
IndexError                                Traceback (most recent call last)
<ipython-input-58-550cc98888dc> in <module>()
----> 1 get_ipython().run_cell_magic('time', '', "data.to_csv('../data/hashtags_binarized.csv', index=False)")

/usr/local/lib/python3.6/site-packages/IPython/core/interactiveshell.py in run_cell_magic(self, magic_name, line, cell)
   2113             magic_arg_s = self.var_expand(line, stack_depth)
   2114             with self.builtin_trap:
-> 2115                 result = fn(magic_arg_s, cell)
   2116             return result
   2117 

<decorator-gen-59> in time(self, line, cell, local_ns)

/usr/local/lib/python3.6/site-packages/IPython/core/magic.py in <lambda>(f, *a, **k)
    186     # but it's overkill for just that one bit of state.
    187     def magic_deco(arg):
--> 188         call = lambda f, *a, **k: f(*a, **k)
    189 
    190         if callable(arg):

/usr/local/lib/python3.6/site-packages/IPython/core/magics/execution.py in time(self, line, cell, local_ns)
   1179         if mode=='eval':
   1180             st = clock2()
-> 1181             out = eval(code, glob, local_ns)
   1182             end = clock2()
   1183         else:

<timed eval> in <module>()

/usr/local/lib/python3.6/site-packages/pandas/core/frame.py in to_csv(self, path_or_buf, sep, na_rep, float_format, columns, header, index, index_label, mode, encoding, compression, quoting, quotechar, line_terminator, chunksize, tupleize_cols, date_format, doublequote, escapechar, decimal)
   1381                                      doublequote=doublequote,
   1382                                      escapechar=escapechar, decimal=decimal)
-> 1383         formatter.save()
   1384 
   1385         if path_or_buf is None:

/usr/local/lib/python3.6/site-packages/pandas/formats/format.py in save(self)
   1473                 self.writer = csv.writer(f, **writer_kwargs)
   1474 
-> 1475             self._save()
   1476 
   1477         finally:

/usr/local/lib/python3.6/site-packages/pandas/formats/format.py in _save(self)
   1574                 break
   1575 
-> 1576             self._save_chunk(start_i, end_i)
   1577 
   1578     def _save_chunk(self, start_i, end_i):

/usr/local/lib/python3.6/site-packages/pandas/formats/format.py in _save_chunk(self, start_i, end_i)
   1588                                   decimal=self.decimal,
   1589                                   date_format=self.date_format,
-> 1590                                   quoting=self.quoting)
   1591 
   1592             for col_loc, col in zip(b.mgr_locs, d):

/usr/local/lib/python3.6/site-packages/pandas/core/internals.py in to_native_types(self, slicer, na_rep, quoting, **kwargs)
    594         values = self.values
    595         if slicer is not None:
--> 596             values = values[:, slicer]
    597         mask = isnull(values)
    598 

/usr/local/lib/python3.6/site-packages/pandas/sparse/array.py in __getitem__(self, key)
    399             return self._get_val_at(key)
    400         elif isinstance(key, tuple):
--> 401             data_slice = self.values[key]
    402         else:
    403             if isinstance(key, SparseArray):

IndexError: too many indices for array

df.to_dense().to_csv("name.csv", index = False, sep=',', encoding='utf-8')