Is python 3.6 required?

I hit syntax error on dl1 lesson 3, on the use of f string format.
Is this new python feature only available in 3.6? Is there a way to run with python 3.5? The reason I may want to say in 3.5 is 'cos I am trying to run them inside a tensorflow docker, and I have other trouble trying to install 3.6 there.

Yeah, f-string is the new feature in 3.6, try to change them to .format() or use concatenation instead.

Thanks. I moved to use docker exactly to avoid this sort of editing. I am trying this docker:


which seemed to have lot of ML/DL stuff installed and with python 3.6. So far, it looks great. I am able to pip install all the requirements, and not hitting errors so far.

Everything is good till running this cell:

for df in (joined,joined_test):
df[“Promo2Since”] = pd.to_datetime(df.apply(lambda x: Week(
x.Promo2SinceYear, x.Promo2SinceWeek).monday(), axis=1).astype(pd.datetime))
df[“Promo2Days”] = df.Date.subtract(df[“Promo2Since”]).dt.days

The kernel just died. Not sure how to debug. Is this a very memory intensive steps? I tried running the notebook in google colab and was able to return without kernel dying.

It will be great if someone can help me debug why the kernel got killed in this cell when i ran this in a docker container on my laptop.