[Python] How to efficiently categorize column in Pandas (no pd.cut)?

I have a large dataframe (millions of rows) and I want to categorize a numerical column into ordered labels, but I want to give names only to some specific values and put all other values into the "transient" category. Using pd.cut looks cumbersome and tricky in some cases.
Is there an efficient way to get the same result of the example code?

import pandas as pd

MODES = [0, 9, 15, 25, 40]
CATS = 'B BC/2 BC AB ABC'.split()

# Build categorical type
my_cats = pd.CategoricalDtype(categories=['transient', 'B', 'BC/2', 'BC', 'AB', 'ABC'], ordered=True)

# The data to be categorized
df = pd.DataFrame(data=[-3, 1.99, 0, 3, 9, 12, 15, 17, 24.9999999, 25, 25.000000001, 34, 40-1e-13, 40], columns=['val'])

# Building...
df['cat'] = df['val'].replace(dict(zip(MODES, CATS)))
c_ = ~df['cat'].isin(CATS)
df.loc[c_, 'cat'] = 'transient'
df['cat'] = df['cat'].astype(my_cats)

# What I want
print(df)

Continue reading...