Python - 将 object 格式的纪元转换为日期时间

Question

I'm pulling data from API that gives me date values in epoch in milliseconds, but the values are in this format:

I am trying to convert these to datetime with pd.to_datetime(df1.loc[:, 'Curve Value'], unit='ms') , but am getting the error

ValueError: non convertible value 1,609,459,200,000.0000000000 with the unit 'ms'

I then tried to format the column into float with df1["Curve Value"] = df["Curve Value"].astype(float) , but get then get this error

ValueError: could not convert string to float: '1,609,459,200,000.0000000000'

I've also tried various ways to remove the commas and convert to float, but get errors trying that, as well.

Answer 1

A bit unwieldy and requires importing datetime from datetime but it works as far as I can see.

df['Real Time'] = df['Curve Value'].apply(lambda t: datetime.fromtimestamp(int(''.join(t.split(',')[:4]))))