Issue in approach 5

[SandeepSince1997]

In approach 5 code, for the function clean_data(inputDF, ....)
function body is given, but function call is not there. What should be the function call and its parameters

[rupav]

This function is to create the stoplist, based on different flags, tf1, stops_bool and tf_high.
Basically,

• stops_bool decides if we want to enter the common STOPWORDS provided in NLP lib used.
• tf_high is used to decide if we want to update stops (stopwords list) using those words which are too common in the given corpus, and thus cant be used for classification.
• updates is a list which of Stopwords, you want to give explicity.
• *tf_high_args is variable argument list, which is directly passed to helper function create_tf_high_stoplist as its argument list. This way I can easily change argument list of create_tf_high_stoplist function without other functions.

After the stops are created, you just CountVectorizer and n_grams approach.

e.g. of the function call would be,

clean_data(df, True, True, True, ['hi'], pos_counts, neg_counts, 1000)

assuming pos_counts, and neg_counts defined above in the code is used.

[SandeepSince1997]

Thank you rupav. I have one more doubt,
In the same function, you have read the file "TF1_cleaned_input.p".
It means there is some data in the file. Where, when and what data is stored in that file. I don't find any line writing data into that file.