Create app.py

app.py

@@ -0,0 +1,97 @@
+import streamlit as st
+st.set_page_config(layout="wide")
+import uuid
+from pathlib import Path
+import os
+import pandas as pd
+import json
+import docx
+# import openai
+# from langchain.chat_models import ChatOpenAI
+# from langchain.prompts import PromptTemplate
+
+st.title("Passage Name with Gen-AI")
+os.environ["OPENAI_API_KEY"] = os.environ["OPEN_API_KEY"]
+from openai import OpenAI
+client = OpenAI()
+list_dict = [] 
+LIMIT_WORDS_PARA = 7
+
+
+
+# Function to guess a single topic for a given text
+def guess_topic(clause_text):
+    response = client.chat.completions.create(
+      model="gpt-4",
+      messages=[
+        {
+          "role": "system",
+          "content": "You are a legal contract assistant. Find type or title of the paragraph, e.g. Governing Law, Confidentiality etc."
+        },
+        {
+          "role": "user",
+          "content": "THIS AGREEMENT SHALL BE GOVERNED BY AND CONSTRUED IN ACCORDANCE WITH THE LAWS OF THE STATE OF NEW YORK."
+        },
+        {
+          "role": "assistant",
+          "content": "Governing Law"
+        },
+        {
+          "role": "user",
+          "content": "The provisions of this Agreement will be held in strictest confidence by you and the Company and will not be publicized or disclosed in any manner whatsoever; provided, however, that: (a) you may disclose this Agreement to your immediate family; (b) the parties may disclose this Agreement in confidence to their respective attorneys, accountants, auditors, tax preparers, and financial advisors; (c) the Company may disclose this Agreement as necessary to fulfill standard or legally required corporate reporting or disclosure requirements; and (d) the parties may disclose this Agreement insofar as such disclosure may be necessary to enforce its terms or as otherwise required by law."
+        },
+        {
+          "role": "assistant",
+          "content": "Confidentiality"
+        },
+        {
+          "role": "user",
+          "content": clause_text
+        },
+      ],
+      temperature=0,
+      max_tokens=10,
+      top_p=1,
+      frequency_penalty=0,
+      presence_penalty=0
+    )
+    return(response)
+    
+with st.form("my_form"):
+    multi = '''
+    
+    1. Upload many files (DOCX) Only
+    
+    2. Press Calculate and Get Name of Clauses
+    
+    '''
+    st.markdown(multi)
+    uploaded_files = st.file_uploader('Upload your files',accept_multiple_files=True)
+    for f in uploaded_files:
+        if True:
+            save_path = Path(os.getcwd(), f.name)
+            with open(save_path, mode='wb') as w:
+                w.write(f.getvalue())
+            doc = docx.Document(save_path)
+            paragraphs = doc.paragraphs
+            for para in paragraphs:
+                text = para.text
+                words = text.split()
+                if len(words) > LIMIT_WORDS_PARA:
+                    list_dict.append({"file":f.name, "example":text})  
+    ###### 
+    print('len(list_dict)',len(list_dict))
+    df_new_trimmed = pd.DataFrame(list_dict)
+    ######
+    submitted = st.form_submit_button("Calculate")
+    if submitted and (uploaded_files is not None):
+        topics_list = []
+        for doc in list(df_new_trimmed['example']):
+            tmp_keywords_custom = guess_topic(doc).choices[0].message.content
+            topics_list.append(tmp_keywords_custom)
+        df_new_trimmed['Predicted_Clause'] = topics_list
+
+        print(topics_list[0])
+
+        st.dataframe(df_new_trimmed)
+        st.write('-----')