UNIX conversion to datetime either mismatch or returns 1970-01-01

See more:

Python3.6

DataFrame

I'm trying to convert a Pandas DataFrame column from UNIX to Datetime, but I either get a mismatch error or the new dates are all 1970-01-01... I'm stuck for hours not knowing how to fix this...

What I have tried:

When I look at how my UNIX looks like in datetime:
<pre>mydatetime = datetime.fromtimestamp(1618185600000 // 1000, tz=tzutc())
print('mydatetime',mydatetime)

I get:

Quote:

mydatetime 2021-04-12 00:00:00+00:00

So when I use the conversion function:

df2 = pd.to_datetime(df1['ds'].astype(str), format='%Y-%m-%d %H:%M:%S+%f:%Z')

I get:

Quote:

ValueError: time data '1618185600000' does not match format '%Y-%m-%d %H:%M:%S+%fZ' (match)

So, when I use the lazy road:

#df2 = pd.to_datetime(df1['ds'], unit='ns')

The results are:

Quote:

DATAFRAMED MARKET TAILED: 86 1970-01-01 00:27:05.616000
87 1970-01-01 00:27:05.702400
88 1970-01-01 00:27:05.788800
89 1970-01-01 00:27:05.875200
90 1970-01-01 00:27:05.886197
Name: ds, dtype: datetime64[ns]