Takuya UESHIN

Seems #726 is related to this issue.

Hi @lamesjaidler, I'm afraid I don't think the comparison above is what you want. I can see a couple of problems there: ### the benchmarks above don't measure the execution...

@HyukjinKwon > If performing Koalas' transpose is expensive, we should switch it to pandas'. The results to transpose is small so I wonder how it makes it slow though. I...

Koalas doesn't support the custom type as a return type from UDFs. How about: ```py >>> kdf['y_vectorized_lambdas'] = kdf.apply(lambda df: mul_func(df).myfunc(), axis=1) >>> kdf col1 col2 y_vectorized_lambdas 0 2 3...

According to the error message, I guess you are using DBR 6.x. If so, you should do `spark.conf.set("spark.sql.execution.arrow.enabled", "false")` instead of `spark.conf.set("spark.sql.execution.arrow.pyspark.enabled", "false")` if you want to disable Arrow optimization.

Could you share a simple example to reproduce the error? The examples in the document seems working finely. ```py >>> s = ks.Series(list('abca')) >>> ks.get_dummies(s) a b c 0 1...

Hi @darshats, Thanks for trying Koalas. Unfortunately Koalas doesn't support `np.timedelta64`, not only in type hints but as a whole, so far. We'll add a documentation about supported types soon....

Here's two things: - read files over `http` protocol - so far, Koalas relies on Spark DataFrameReader which doesn't support `http` protocol. We need some workaround to support it. -...

As for the `head` behavior, we added a config to perform natural ordering when `head` (#1231). If you turn `compute.ordered_head` to `True`, you will see the same result as pandas'...

You can use [`DataFrame.spark.coalesce`](https://koalas.readthedocs.io/en/latest/reference/api/databricks.koalas.DataFrame.spark.coalesce.html) or [`DataFrame.spark.repartition`](https://koalas.readthedocs.io/en/latest/reference/api/databricks.koalas.DataFrame.spark.repartition.html) to control the file number. ```py df.spark.coalesce(num_files).to_parquet(...) ``` For reading, you don't need to append/concatenate the files by yourself. Koalas will automatically do it...