Perform the same text processing with Pandas. From a Notebook. And save results into a sqlite3 database.
To do that, you'll need to think about the dataframe you need to create. What columns does it have?
How do you make a relationship between the document data (the name, the word count, the lines and char counts) and the frequencies.
Probably two tables, word_freq and letter_freq? Or can you think of a better way to turn the data into a schema?