dtype for DateType should be np.dtype("object").

[ueshin]

dtype for DateType should be np.dtype("object").

>>> import pandas as pd
>>> import datetime
>>> pdf = pd.DataFrame({'dt': [datetime.date(2020,1,1),datetime.date(2020,1,2)], 'ts':[datetime.datetime(2020,1,1,0,0,0), datetime.datetime(2020,1,2,0,0,0)]})
>>> pdf.dtypes
dt            object
ts    datetime64[ns]
dtype: object

whereas:

>>> ks.from_pandas(pdf).dtypes
dt    datetime64[ns]
ts    datetime64[ns]
dtype: object

[codecov-io]

Codecov Report

Merging #1447 into master will decrease coverage by 2.07%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #1447      +/-   ##
==========================================
- Coverage   94.00%   91.93%   -2.08%     
==========================================
  Files          36       36              
  Lines        8342     8341       -1     
==========================================
- Hits         7842     7668     -174     
- Misses        500      673     +173     

Impacted Files	Coverage Δ
databricks/koalas/typedef/typehints.py	81.08% <100.00%> (-0.17%)	⬇️
databricks/koalas/usage_logging/__init__.py	24.32% <0.00%> (-70.28%)	⬇️
databricks/koalas/usage_logging/usage_logger.py	50.00% <0.00%> (-50.00%)	⬇️
databricks/koalas/__init__.py	78.84% <0.00%> (-13.47%)	⬇️
databricks/conftest.py	88.67% <0.00%> (-7.55%)	⬇️
databricks/koalas/frame.py	92.54% <0.00%> (-3.02%)	⬇️
databricks/koalas/namespace.py	84.40% <0.00%> (-1.62%)	⬇️
databricks/koalas/plot.py	93.33% <0.00%> (-0.96%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 7ea6739...86f5d65. Read the comment docs.