Cannot read h5ad file

[chriscainx]

>>> adata.write("./result.h5ad")
>>> bdata = sc.read("./result.h5ad)

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-15-d90a365327a0> in <module>()
      1 adata.write(“./results.h5ad”)
----> 2 bdata = sc.read(“./results.h5ad")

/fastdata/chris/bin/anaconda3/lib/python3.6/site-packages/anndata/readwrite/read.py in read_h5ad(filename, backed)
    347         # load everything into memory
    348         d = _read_h5ad(filename=filename)
--> 349         return AnnData(d)
    350 
    351 

/fastdata/chris/bin/anaconda3/lib/python3.6/site-packages/anndata/base.py in __init__(self, X, obs, var, uns, obsm, varm, raw, dtype, single_col, filename, filemode, asview, oidx, vidx)
    632                 obsm=obsm, varm=varm, raw=raw,
    633                 dtype=dtype, single_col=single_col,
--> 634                 filename=filename, filemode=filemode)
    635 
    636     def _init_as_view(self, adata_ref, oidx, vidx):

/fastdata/chris/bin/anaconda3/lib/python3.6/site-packages/anndata/base.py in _init_as_actual(self, X, obs, var, uns, obsm, varm, raw, dtype, single_col, filename, filemode)
    741                 raise ValueError(
    742                     'If `X` is a dict no further arguments must be provided.')
--> 743             X, obs, var, uns, obsm, varm, raw = self._from_dict(X)
    744 
    745         # init from AnnData

/fastdata/chris/bin/anaconda3/lib/python3.6/site-packages/anndata/base.py in _from_dict(ddata)
   1591                     d_true_keys['obs'][k_stripped] = pd.Categorical.from_codes(
   1592                         codes=d_true_keys['obs'][k_stripped].values,
-> 1593                         categories=v)
   1594                 if k_stripped in d_true_keys['var']:
   1595                     d_true_keys['var'][k_stripped] = pd.Categorical.from_codes(

/fastdata/chris/bin/anaconda3/lib/python3.6/site-packages/pandas/core/categorical.py in from_codes(cls, codes, categories, ordered)
    616                 "codes need to be convertible to an arrays of integers")
    617 
--> 618         categories = CategoricalDtype._validate_categories(categories)
    619 
    620         if len(codes) and (codes.max() >= len(categories) or codes.min() < -1):

/fastdata/chris/bin/anaconda3/lib/python3.6/site-packages/pandas/core/dtypes/dtypes.py in _validate_categories(categories, fastpath)
    324                 raise ValueError('Categorial categories cannot be null')
    325 
--> 326             if not categories.is_unique:
    327                 raise ValueError('Categorical categories must be unique')
    328 

pandas/_libs/properties.pyx in pandas._libs.properties.cache_readonly.__get__()

/fastdata/chris/bin/anaconda3/lib/python3.6/site-packages/pandas/core/indexes/base.py in is_unique(self)
   1316     def is_unique(self):
   1317         """ return if the index has unique values """
-> 1318         return self._engine.is_unique
   1319 
   1320     @property

pandas/_libs/properties.pyx in pandas._libs.properties.cache_readonly.__get__()

/fastdata/chris/bin/anaconda3/lib/python3.6/site-packages/pandas/core/indexes/base.py in _engine(self)
   1595     def _engine(self):
   1596         # property, for now, slow to look up
-> 1597         return self._engine_type(lambda: self._values, len(self))
   1598 
   1599     def _validate_index_level(self, level):

/fastdata/chris/bin/anaconda3/lib/python3.6/site-packages/pandas/core/indexes/base.py in __len__(self)
    559         return the length of the Index
    560         """
--> 561         return len(self._data)
    562 
    563     def __array__(self, dtype=None):

TypeError: len() of unsized object

[falexwolf]

Sorry, that's probably due to the fact that your annotations contain a column with a categorical data and a single category. We'll solve this bug in the next version of anndata, just a couple of days.

As workaround, remove the annotation with a single cateogry.

[chriscainx]

thank you. please close this issue

[gtca]

It seems that the issue still can be reproduced in scanpy v1.4:

import pandas as pd
import scanpy as sc
pd.__version__  # => '0.23.4'
sc.__version__  # => '1.4'

adata = sc.datasets.pbmc68k_reduced()
adata.obs["single_cat"] = 1
adata.obs['single_cat'] = pd.Categorical(adata.obs['single_cat'])
adata.write('/tmp/adata.h5ad')

sc.read('/tmp/adata.h5ad')  # => ValueError: codes need to be convertible to an arrays of integers

[flying-sheep]

do you also have the newest anndata version?

[sophiamaedler]

I am also experiencing this issue.

Running the following code:

import pandas as pd
import scanpy as sc
import anndata

print(pd.__version__)
print(sc.__version__)
print(anndata.__version__)
adata = sc.datasets.pbmc68k_reduced()
adata.obs["single_cat"] = 1
adata.obs['single_cat'] = pd.Categorical(adata.obs['single_cat'])
adata.write('/tmp/adata.h5ad')
sc.read('/tmp/adata.h5ad')

Returns this error message:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-5-adde38d13544> in <module>
----> 1 sc.read('/tmp/adata.h5ad')

/usr/local/anaconda3/envs/diffxpy/lib/python3.6/site-packages/scanpy/readwrite.py in read(filename, backed, sheet, ext, delimiter, first_column_names, backup_url, cache, **kwargs)
     95             filename, backed=backed, sheet=sheet, ext=ext,
     96             delimiter=delimiter, first_column_names=first_column_names,
---> 97             backup_url=backup_url, cache=cache, **kwargs,
     98         )
     99     # generate filename and read to dict

/usr/local/anaconda3/envs/diffxpy/lib/python3.6/site-packages/scanpy/readwrite.py in _read(filename, backed, sheet, ext, delimiter, first_column_names, backup_url, cache, suppress_cache_warning, **kwargs)
    497     if ext in {'h5', 'h5ad'}:
    498         if sheet is None:
--> 499             return read_h5ad(filename, backed=backed)
    500         else:
    501             logg.debug(f'reading sheet {sheet} from file {filename}')

/usr/local/anaconda3/envs/diffxpy/lib/python3.6/site-packages/anndata/readwrite/read.py in read_h5ad(filename, backed, chunk_size)
    445     else:
    446         # load everything into memory
--> 447         constructor_args = _read_args_from_h5ad(filename=filename, chunk_size=chunk_size)
    448         X = constructor_args[0]
    449         dtype = None

/usr/local/anaconda3/envs/diffxpy/lib/python3.6/site-packages/anndata/readwrite/read.py in _read_args_from_h5ad(adata, filename, mode, chunk_size)
    500     if not backed:
    501         f.close()
--> 502     return AnnData._args_from_dict(d)
    503 
    504 

/usr/local/anaconda3/envs/diffxpy/lib/python3.6/site-packages/anndata/core/anndata.py in _args_from_dict(ddata)
   2182                         d_true_keys[ann][k_stripped] = pd.Categorical.from_codes(
   2183                             codes=d_true_keys[ann][k_stripped].values,
-> 2184                             categories=v,
   2185                         )
   2186                 k_to_delete.append(k)

/usr/local/anaconda3/envs/diffxpy/lib/python3.6/site-packages/pandas/core/arrays/categorical.py in from_codes(cls, codes, categories, ordered, dtype)
    638         dtype = CategoricalDtype._from_values_or_dtype(categories=categories,
    639                                                        ordered=ordered,
--> 640                                                        dtype=dtype)
    641         if dtype.categories is None:
    642             msg = ("The categories must be provided in 'categories' or "

/usr/local/anaconda3/envs/diffxpy/lib/python3.6/site-packages/pandas/core/dtypes/dtypes.py in _from_values_or_dtype(cls, values, categories, ordered, dtype)
    322             # Note: This could potentially have categories=None and
    323             # ordered=None.
--> 324             dtype = CategoricalDtype(categories, ordered)
    325 
    326         return dtype

/usr/local/anaconda3/envs/diffxpy/lib/python3.6/site-packages/pandas/core/dtypes/dtypes.py in __init__(self, categories, ordered)
    224 
    225     def __init__(self, categories=None, ordered=None):
--> 226         self._finalize(categories, ordered, fastpath=False)
    227 
    228     @classmethod

/usr/local/anaconda3/envs/diffxpy/lib/python3.6/site-packages/pandas/core/dtypes/dtypes.py in _finalize(self, categories, ordered, fastpath)
    333         if categories is not None:
    334             categories = self.validate_categories(categories,
--> 335                                                   fastpath=fastpath)
    336 
    337         self._categories = categories

/usr/local/anaconda3/envs/diffxpy/lib/python3.6/site-packages/pandas/core/dtypes/dtypes.py in validate_categories(categories, fastpath)
    502         if not fastpath and not is_list_like(categories):
    503             msg = "Parameter 'categories' must be list-like, was {!r}"
--> 504             raise TypeError(msg.format(categories))
    505         elif not isinstance(categories, ABCIndexClass):
    506             categories = Index(categories, tupleize_cols=False)

TypeError: Parameter 'categories' must be list-like, was 1

I am running the following versions:

0.24.2 #pandas
1.4.4.post1 #scanpy
0.6.22.post1 #anndata