mgaughan/24_deb_pkg_gov
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

updated topic model distributions

Browse Source
This commit is contained in:
Matthew Gaughan 2024-06-26 09:21:50 -05:00
parent 8f1d4adc1e
commit 979dd9ccc6
1 changed files with 16 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
text_analysis/topicModel.py
View File

@ -113,7 +113,9 @@ def best_lda_model(data_vectorized, vocab):
#lda = joblib.load('0509_lda.jl')
return id_topic
def get_most_prevalent(distributions, documents):
def get_most_prevalent(vect_documents, documents):
lda = joblib.load('0509_readme_lda.jl')
distributions = lda.transform(vect_documents)
most_prevalent = {0: [0, ""],1: [0, ""], 2: [0, ""], 3: [0, ""], 4: [0, ""], 5: [0, ""], 6: [0, ""], 7: [0, ""]}
for i, topic_distribution in enumerate(distributions):
for j in range(8):
@ -123,7 +125,8 @@ def get_most_prevalent(distributions, documents):
return most_prevalent
def prevalent_topics(vect_documents, file_list):
lda = joblib.load('0509_readme_lda.jl')
#lda = joblib.load('0509_readme_lda.jl')
lda = joblib.load('0514_contrib_lda.jl')
distributions = lda.transform(vect_documents)
#figuring out what the max distribution is and then figuring out the mode
top_topic = []
@ -137,10 +140,11 @@ def prevalent_topics(vect_documents, file_list):
else:
count_of_multiple += 1
topic_arrays.append(topic_distribution)
#most_frequent(top_topic)
most_frequent(top_topic)
print(count_of_multiple)
df = pd.DataFrame(topic_arrays)
#print(df.sort_values(by=['0']).head(5))
'''
for i in range(8):
print("-----------------------Topic " + str(i) + " --------------------------------")
top5 = df.nlargest(10, i)
@ -153,12 +157,13 @@ def prevalent_topics(vect_documents, file_list):
print(bottom5)
for index in bottom_indices:
print(file_list[index])
#averages = df.mean()
#print(averages)
'''
averages = df.mean()
print(averages)
def most_frequent(topic_prevalence):
most_frequent_array = []
for j in range(8):
for j in range(4):
topic = mode(topic_prevalence)
most_frequent_array.append(topic)
topic_prevalence = [i for i in topic_prevalence if i != topic]
@ -167,9 +172,9 @@ def most_frequent(topic_prevalence):
if __name__ == "__main__":
readme_directory = "/data/users/mgaughan/kkex/time_specific_files/readme3"
contributing_directory = "/data/users/mgaughan/kkex/time_specific_files/partitioned_contributing/p2"
listed_corpus, wordcounts, wordlengths, file_list = get_data_from_dir(readme_directory)
#readme_directory = "/data/users/mgaughan/kkex/time_specific_files/dwo_partitioned_readme/p2"
contributing_directory = "/data/users/mgaughan/kkex/time_specific_files/dwo_partitioned_contributing/p2"
listed_corpus, wordcounts, wordlengths, file_list = get_data_from_dir(contributing_directory)
print("Mean wordcount: ", mean(wordcounts))
print("Median wordcount: ", median(wordcounts))
print("Mean wordlength: ", mean(wordlengths))
@ -184,11 +189,11 @@ if __name__ == "__main__":
)
data_vectorized = vectorizer.fit_transform(lemmatized_corpus)
'''
vectorizer = joblib.load('readme_vectorizer.jl')
vectorizer = joblib.load('contrib_vectorizer.jl')
data_vectorized = vectorizer.transform(lemmatized_corpus)
#lda_model_identification(data_vectorized)
#topic_distributions = best_lda_model(data_vectorized, vectorizer.get_feature_names_out())
#get_most_prevalent(topic_distributions, file_list)
#get_most_prevalent(data_vectorized, file_list)
prevalent_topics(data_vectorized, file_list)