def merge_topics(request):
form = forms.MergeTopic(request.POST or None)
if request.method == "POST":
if form.is_valid():
topic_number_1 = form.cleaned_data['topic_number_1']
topic_number_2 = form.cleaned_data['topic_number_2']
print("Topic num is: ", topic_number_1)
print("Topic num is: ", topic_number_2)
print(trial['W'])
print(trial['H'])
m = topic_modeling(new_dict['_A'], content['max_key'],
new_dict['_terms'], new_dict['final'])
m.nmf_modeling()
w, h = m.fit_transform_merge(topic_number_1, topic_number_2,
trial['W'], trial['H'])
df = m.top_three_keywords()
_df2, _topic_word = m.documents_per_topic(w, df)
new_df = m.remove_row(_topic_word)
m.frequency_plot(new_df)
m.percentage_plot(new_df)
m.final_output(new_df, w)
m.printPFsInPairs()
m.plot_top_words(3, 'Per topic top 3 keywords')
finalTableData = pd.read_csv(
'../static/assets/themes_keywords.csv')
content['columns'] = finalTableData.columns
content['rows'] = finalTableData.to_dict('records')
print(finalTableData.head())
return render(request, 'basic.html', content)
def split_topic(request):
splitForm = forms.SplitTopic(request.POST or None)
if request.method == 'POST':
print(splitForm)
if splitForm.is_valid():
print('hello yes it works')
topic_number = splitForm.cleaned_data['topic_number']
topic_1_keywords = splitForm.cleaned_data['topic_1_keywords']
topic_2_keywords = splitForm.cleaned_data['topic_2_keywords']
print("Topic Number is: ", topic_number)
print("Topic_1_keywords: ", topic_1_keywords)
print("other keywords: ", topic_2_keywords)
m = topic_modeling(new_dict['_A'], content['max_key'],
new_dict['_terms'], new_dict['final'])
H1, W1 = m.nmf_modeling()
# print(W)
df = m.top_three_keywords()
_df2, _topic_word = m.documents_per_topic(W1, df)
m.frequency_plot(_topic_word)
m.percentage_plot(_topic_word)
output = m.final_output(_topic_word, W1)
m.split_topic(topic_number, topic_1_keywords, topic_2_keywords,
output)
return render(request, 'basic.html', content)
def change_topic(request):
form = forms.ChangeTopic()
if request.method == "POST":
form = forms.ChangeTopic(request.POST)
if form.is_valid():
new_topic_number = form.cleaned_data['change_topic']
print("Change the number of topics: ", new_topic_number)
print(type(new_topic_number))
topic_number = int(new_topic_number)
print(type(topic_number))
print(new_dict['_A'])
print(new_dict['_terms'])
# print("***")
print(new_dict['final'])
m = topic_modeling(new_dict['_A'], topic_number,
new_dict['_terms'], new_dict['final'])
H1, W1 = m.nmf_modeling()
# print(W)
df = m.top_three_keywords()
_df2, _topic_word = m.documents_per_topic(W1, df)
m.frequency_plot(_topic_word)
m.percentage_plot(_topic_word)
output = m.final_output(_topic_word, W1)
m.printPFsInPairs()
m.plot_top_words(3, 'Per topic top 3 keywords')
content["max_key"] = topic_number
finalTableData = pd.read_csv(
'../static/assets/themes_keywords.csv')
content['columns'] = finalTableData.columns
content['rows'] = finalTableData.to_dict('records')
return render(request, 'basic.html', {
'rows': content['rows'],
'columns': content['columns']
})
def rename_topic(request):
form = forms.RenameTopic(request.POST or None)
if request.method == "POST":
if form.is_valid():
topic_number = form.cleaned_data['topic_number']
name = form.cleaned_data['name']
print("Topic nume is: ", topic_number)
print("rename is: ", name)
m = topic_modeling(new_dict['_A'], content['max_key'],
new_dict['_terms'], new_dict['final'])
m.rename_topic(topic_number, name, new_dict['theme_keywords'])
finalTableData = pd.read_csv(
'../static/assets/themes_keywords.csv')
content['columns'] = finalTableData.columns
content['rows'] = finalTableData.to_dict('records')
print(finalTableData.head())
return render(request, 'basic.html', content)
def delete_keywords(request):
form = KeyWordDeletionForm(request.POST or None)
if request.method == 'POST':
print(form)
if form.is_valid():
print('hello yes it works')
new_stopwords = form.cleaned_data['delete_keyword']
print("Exclude word", new_stopwords)
n_class = Stopwords(add_stopwords=new_stopwords)
additional = n_class.adding_stopwords()
content['additional_stopwords'] = additional
x = Automate_topic_modeling(content['file'], additional)
new_dict['final'] = x.pre_processing()
x.generate_wordcloud(new_dict['final'])
new_dict['_terms'], new_dict[
'_A'], _raw_documents = x.tfidf_matrix(new_dict['final'])
# topic_models, H, W = x.nmf_model(new_dict['_A'])
# x.build_w2c(_raw_documents)
# term_rankings, max_key, newDict = x.get_coherence(new_dict['_terms'], topic_models)
# plot = x.plot_the_coherence_graph(newDict)
#
m = topic_modeling(new_dict['_A'], content['max_key'],
new_dict['_terms'], new_dict['final'])
H1, W1 = m.nmf_modeling()
trial['H'] = H1
trial['W'] = W1
# print(W)
df = m.top_three_keywords()
_df2, _topic_word = m.documents_per_topic(W1, df)
m.frequency_plot(_topic_word)
m.percentage_plot(_topic_word)
theme_keywords = m.final_output(_topic_word, W1)
m.printPFsInPairs()
m.plot_top_words(3, 'Per topic top 3 keywords')
new_dict['theme_keywords'] = theme_keywords
finalTableData = pd.read_csv(
'../static/assets/themes_keywords.csv')
content['columns'] = finalTableData.columns
content['rows'] = finalTableData.to_dict('records')
content['id'] = finalTableData['Topic_id'].tolist()
content['top10'] = finalTableData['Top_ten_words'].tolist()
return render(request, 'basic.html', content)
def top_responses(request):
form = forms.TopResponses(request.POST or None)
if request.method == "POST":
if form.is_valid():
top_responses = form.cleaned_data['top_responses']
print("top_responses_are: ", top_responses)
m = topic_modeling(new_dict['_A'], content['max_key'],
new_dict['_terms'], new_dict['final'])
H1, W1 = m.nmf_modeling()
df = m.top_three_keywords()
_df2, _topic_word = m.documents_per_topic(W1, df)
m.frequency_plot(_topic_word)
m.percentage_plot(_topic_word)
output = m.final_output(_topic_word, W1, top=top_responses)
finalTableData = pd.read_csv(
'../static/assets/themes_keywords.csv')
content['columns'] = finalTableData.columns
content['rows'] = finalTableData.to_dict('records')
print(finalTableData.head())
return render(request, 'basic.html', content)
def addstopwords(request):
form_2 = forms.Add_stopwords(request.POST)
if request.method == "POST":
print(form_2)
if form_2.is_valid():
new_stopwords = form_2.cleaned_data['add_stopwords']
print(type(new_stopwords))
print("New stopwords is:", new_stopwords)
n_class = Stopwords(add_stopwords=new_stopwords)
additional = n_class.adding_stopwords()
content['additional_stopwords'] = additional
x = Automate_topic_modeling(content['file'], additional)
new_dict['final'] = x.pre_processing()
x.generate_wordcloud(new_dict['final'])
new_dict['_terms'], new_dict[
'_A'], _raw_documents = x.tfidf_matrix(new_dict['final'])
# topic_models, H, W = x.nmf_model(new_dict['_A'])
# x.build_w2c(_raw_documents)
m = topic_modeling(new_dict['_A'], content['max_key'],
new_dict['_terms'], new_dict['final'])
H1, W1 = m.nmf_modeling()
trial['H'] = H1
trial['W'] = W1
# print(W)
df = m.top_three_keywords()
_df2, _topic_word = m.documents_per_topic(W1, df)
m.frequency_plot(_topic_word)
m.percentage_plot(_topic_word)
theme_keywords = m.final_output(_topic_word, W1)
m.printPFsInPairs()
m.plot_top_words(3, 'Per topic top 3 keywords')
new_dict['theme_keywords'] = theme_keywords
finalTableData = pd.read_csv(
'../static/assets/themes_keywords.csv')
content['columns'] = finalTableData.columns
content['rows'] = finalTableData.to_dict('records')
content['id'] = finalTableData['Topic_id'].tolist()
content['top10'] = finalTableData['Top_ten_words'].tolist()
return render(request, 'basic.html', content)
def upload(request):
global content
max_key = int
if request.method == 'POST':
uploaded_file = request.FILES['document']
file_name = uploaded_file.name
print(uploaded_file.name)
print(uploaded_file.size)
fs = FileSystemStorage()
fs.save(uploaded_file.name, uploaded_file)
form_1 = forms.SheetName(request.POST)
if form_1.is_valid():
sheet_name = form_1.cleaned_data['sheetname']
output_file_name = form_1.cleaned_data['output_file_name']
checkbox = request.POST.get('vehicle1')
print("Sheet name name is: ", sheet_name)
print("output file name is: ", output_file_name)
print("Love")
# print("Current date is", current_date)
file_1 = generate_csvfile(file_name, sheet_name)
final_csv = file_1.convert_file()
content['file'] = final_csv
# New class
if (final_csv.empty == False):
x = Automate_topic_modeling(final_csv)
if checkbox == 'on':
print("yes")
x.remove_numbers()
new_dict['final'] = x.pre_processing()
x.generate_wordcloud(new_dict['final'])
new_dict['_terms'], new_dict[
'_A'], _raw_documents = x.tfidf_matrix(new_dict['final'])
topic_models, H, W = x.nmf_model(new_dict['_A'])
x.build_w2c(_raw_documents)
term_rankings, max_key, newDict = x.get_coherence(
new_dict['_terms'], topic_models)
content['max_key'] = max_key
plot = x.plot_the_coherence_graph(newDict)
m = topic_modeling(new_dict['_A'], max_key, new_dict['_terms'],
new_dict['final'])
H1, W1 = m.nmf_modeling()
trial['H'] = H1
trial['W'] = W1
# print(W)
df = m.top_three_keywords()
_df2, _topic_word = m.documents_per_topic(W1, df)
m.frequency_plot(_topic_word)
m.percentage_plot(_topic_word)
theme_keywords = m.final_output(_topic_word, W1)
m.printPFsInPairs()
m.plot_top_words(3, 'Per topic top 3 keywords')
new_dict['theme_keywords'] = theme_keywords
# content = {'max_key': max_key, 'output_file_name': output_file_name}
content['output_file_name'] = output_file_name
finalTableData = pd.read_csv(
'../static/assets/themes_keywords.csv')
content['columns'] = finalTableData.columns
content['rows'] = finalTableData.to_dict('records')
print("columns:")
print(content['columns'])
print("Rows:")
print(content['rows'])
content['id'] = finalTableData['Topic_id'].tolist()
content['top10'] = finalTableData['Top_ten_words'].tolist()
print("F**K")
print((content['top10']))
print(finalTableData.head())
else:
print('ADD CORRECT FILE')
else:
output_file_name = form_1.cleaned_data['output_file_name']
print("output file name is: ", output_file_name)
checkbox = request.POST.get('vehicle1')
print(checkbox)
print("LOve")
# print("Current date is", current_date)
file_1 = generate_csvfile(file_name)
final_csv = file_1.convert_file()
content['file'] = final_csv
# New class
if (final_csv.empty == False):
x = Automate_topic_modeling(final_csv)
if checkbox == 'on':
print("yes")
x.remove_numbers()
new_dict['final'] = x.pre_processing()
x.generate_wordcloud(new_dict['final'])
new_dict['_terms'], new_dict[
'_A'], _raw_documents = x.tfidf_matrix(new_dict['final'])
topic_models, H, W = x.nmf_model(new_dict['_A'])
x.build_w2c(_raw_documents)
term_rankings, max_key, newDict = x.get_coherence(
new_dict['_terms'], topic_models)
plot = x.plot_the_coherence_graph(newDict)
m = topic_modeling(new_dict['_A'], max_key, new_dict['_terms'],
new_dict['final'])
H1, W1 = m.nmf_modeling()
trial['H'] = H1
trial['W'] = W1
# print(W)
df = m.top_three_keywords()
_df2, _topic_word = m.documents_per_topic(W1, df)
m.frequency_plot(_topic_word)
m.percentage_plot(_topic_word)
theme_keywords = m.final_output(_topic_word, W1)
m.printPFsInPairs()
m.plot_top_words(3, 'Per topic top 3 keywords')
new_dict['theme_keywords'] = theme_keywords
# content = {'max_key': max_key, 'output_file_name': output_file_name}
content['max_key'] = max_key
content['output_file_name'] = output_file_name
finalTableData = pd.read_csv(
'../static/assets/themes_keywords.csv')
content['columns'] = finalTableData.columns
content['rows'] = finalTableData.to_dict('records')
content['id'] = finalTableData['Topic_id'].tolist()
content['top10'] = finalTableData['Top_ten_words'].tolist()
print(finalTableData.head())
else:
print('ADD CORRECT FILE')
return render(request, 'basic.html', content)