scholarly360 commited on
Commit
14f9685
·
verified ·
1 Parent(s): 082ecbe

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +15 -2
app.py CHANGED
@@ -12,7 +12,9 @@ from langchain.retrievers import BM25Retriever, EnsembleRetriever
12
  from langchain.schema import Document
13
  from langchain.vectorstores import Chroma
14
  from langchain.embeddings import HuggingFaceEmbeddings
15
-
 
 
16
 
17
  def util_upload_file_and_return_list_docs(uploaded_files):
18
  #util_del_cwd()
@@ -130,10 +132,21 @@ with st.form("my_form"):
130
  list_save_path = []
131
  uploaded_files = st.file_uploader("Choose file(s)", accept_multiple_files=True)
132
  print('uploaded_files ', uploaded_files)
133
- submitted = st.form_submit_button("Index and Calculate")
 
 
 
134
 
135
  if submitted and (uploaded_files is not None):
136
  list_docs, list_save_path = util_upload_file_and_return_list_docs(uploaded_files)
137
  # print('list_docs ' ,list_docs)
138
  # print('list_save_path ' , list_save_path)
139
  passage_documents = util_get_list_page_and_passage(list_docs, list_save_path)
 
 
 
 
 
 
 
 
 
12
  from langchain.schema import Document
13
  from langchain.vectorstores import Chroma
14
  from langchain.embeddings import HuggingFaceEmbeddings
15
+ from setfit import SetFitModel
16
+ # Download from the 🤗 Hub
17
+ clause_model = SetFitModel.from_pretrained("scholarly360/setfit-contracts-clauses")
18
 
19
  def util_upload_file_and_return_list_docs(uploaded_files):
20
  #util_del_cwd()
 
132
  list_save_path = []
133
  uploaded_files = st.file_uploader("Choose file(s)", accept_multiple_files=True)
134
  print('uploaded_files ', uploaded_files)
135
+ submitted = st.form_submit_button("Calculate")
136
+
137
+ my_list_structure = []
138
+ import pandas as pd
139
 
140
  if submitted and (uploaded_files is not None):
141
  list_docs, list_save_path = util_upload_file_and_return_list_docs(uploaded_files)
142
  # print('list_docs ' ,list_docs)
143
  # print('list_save_path ' , list_save_path)
144
  passage_documents = util_get_list_page_and_passage(list_docs, list_save_path)
145
+ for passage_document in passage_documents:
146
+ text = passage_document.page_content
147
+ metadata = passage_document.metadata
148
+ preds = model(text)
149
+ my_list_structure.append({"text": text, "metadata": metadata,"preds":preds })
150
+
151
+ df = pd.DataFrame(my_list_structure)
152
+ df