import streamlit as st import pandas as pd from top2vec import Top2Vec # def append_dict(data): # # Create df with author info # author_df = pd.DataFrame(columns=['Name', 'Affiliation']) # for i, item in enumerate(data): # author_df.at[i, 'Name'] = item['name'] # if item['affiliation'] is not None: # author_df.at[i, 'Affiliation'] = item['affiliation'] # else: # author_df.at[i, 'Affiliation'] = '' # # Find longest affiliation string # max_affiliation = 0 # for item in author_df['Affiliation']: # if len(item) > max_affiliation: # max_affiliation = len(item) # # Add space to affiliation strings to match longest string # for i, item in enumerate(author_df['Affiliation']): # author_df.at[i, 'Affiliation'] = item + ' ' * (max_affiliation - len(item)) # # Change index to start at 1 # author_df.index = author_df.index + 1 # return author_df @st.cache_resource def load_model(): name = '2024-02-14_22_19_09_reduced__model_medline0[test,mincount_250,topicmergedelta_0.15,ngram_True,ngramvocab_args_250_.4_80M_npmi_connector].json' with open(name, 'rb') as f: return Top2Vec.load(f) @st.cache_data def load_data(): df = pd.read_pickle('medline_0_df_nlp.pkl.bz2', compression='bz2') return df # Notification of loading start_container = st.empty() if 'first' not in st.session_state: st.session_state['first'] = True if st.session_state['first'] == True: start_container.text('Loading model and data... This may take two to three minutes on the initial run.') st.session_state['first'] = False # Load the model and data model = load_model() df = load_data() # Remove loading notification start_container.empty() # User interface with st.sidebar.form('Explore Documents', clear_on_submit=False): st.write('Input a search term(s) separated by a space and receive the documents that are most semantical similar.') txt_ed = st.text_area('Keyword(s) or Phrase(s)') num_ed = st.number_input('Number of documents to return', min_value=1, max_value=50, value=10) # Default value is 10 submitted_ed = st.form_submit_button("Submit") if submitted_ed: txt_ed_list = txt_ed.split() doc_scores, doc_ids = model.search_documents_by_keywords(txt_ed_list, num_docs=num_ed) for i, doc in enumerate(doc_ids): st.header(f'{df.at[doc, "TI"]}') with st.expander('See more'): # st.write('Authors:', f'{df.at[doc, "AU"]}') st.write('Authors:', ','.join(df.at[doc, "AU"]).replace(',', ', ')) # st.dataframe(append_dict(df.iat[doc, 5])) st.write('Year:', f'{df.at[doc, "DP"]}') st.markdown(f'PMUID: [{df.at[doc, "PMID"]}](https://pubmed.ncbi.nlm.nih.gov/{df.at[doc, "PMID"]})') st.write('Journal:', f'*{df.at[doc, "JT"]}*') st.write('Journal Location:', f'{df.at[doc, "PL"]}') st.write('Cosine Similarity:', str(doc_scores[i])) st.write('Abstract:', f'{df.at[doc,"AB"]}') st.write('Keywords:', '|'.join(df.at[doc, "MH"]).replace('|', ' | '))