pr19

app.py

@@ -1,14 +1,6 @@
 import streamlit as st
-import time
+from utils import generate_email
 
-# def submit_callback():
-    # st.session_state.clicked = False
-    # handle the user_input here
-    # st.write("Product: " + st.session_state.product)
-    # st.write("Gender: " + st.session_state.gender) 
-    # st.write("Profession: " + st.session_state.profession) 
-    # st.write("Hobby: "+ st.session_state.hobby)
-    
     
 col1, col2=st.columns(2)
 with col1:
@@ -18,18 +10,24 @@ with col1:
             # product
             product=st.text_input("product", key="product")
             # gender
-            gender=st.radio("gender", ["male", "female", "Non-binary"], key="gender")
+            gender=st.radio("gender", ["male", "female", "non-binary"], key="gender")
             # profession
             profession=st.text_input("profession", key="profession")
             # hobby
             hobby=st.text_input("hobby", key="hobby")
             # Every form must have a submit button.
-            # submitted = st.form_submit_button(label='Submit', on_click=submit_callback)
             submitted = st.form_submit_button(label='Submit')
+            # if submitted: 
+            #      response = generate_email(st.session_state.product, st.session_state.gender, st.session_state.profession, st.session_state.hobby)
+            #      st.session_state.email=response
+                 
 with col2:
         if submitted:
             with st.spinner('Wait for it...'):
-                time.sleep(5)
+                response = generate_email(st.session_state.product, st.session_state.gender, st.session_state.profession, st.session_state.hobby)
+                st.session_state.email=response
+                print("st.session_state.email")
+                print(st.session_state.email)
                 with st.container(height=500):
                     placeholder = st.empty()
                     # Replace the chart with several elements:
@@ -38,6 +36,7 @@ with col2:
                         st.write("Gender: " + st.session_state.gender) 
                         st.write("Profession: " + st.session_state.profession) 
                         st.write("Hobby: "+ st.session_state.hobby)
+                        st.write("Email: "+ st.session_state.email)
 
 if 'clicked' not in st.session_state:
         st.session_state.clicked = False
@@ -53,317 +52,7 @@ if st.session_state.clicked:
     del st.session_state.gender
     del st.session_state.profession
     del st.session_state.hobby
+    del st.session_state.email
     del st.session_state.clicked
     if submitted:
         placeholder.empty()
-
-
-
-# Clear all those elements:
-# placeholder.empty()
-        
-# if clear:
-#     del st.session_state.product
-#     del st.session_state.gender
-#     del st.session_state.profession
-#     del st.session_state.hobby
-#     placeholder.empty()
-
-# container1 = st.container(border=True)
-
-# col1, col2 = st.columns(2)
-# with col1:
-#     container1 = col1.container(height=3000, )
-
-# with col2:
-#     container2 = col2.container(height=300)
-
-# placeholder = st.empty()
-
-
-# placeholder = st.empty()
-# # Create two columns
-# col1, col2 = st.columns(2)
-
-# # Place a form in each column
-# with col1:
-# def submit_callback():
-#     st.write("Product: " + st.session_state.product)
-#     st.write("Gender: " + st.session_state.gender) 
-#     st.write("Profession: " + st.session_state.profession) 
-#     st.write("Hobby: "+ st.session_state.hobby)
-
-# def delete_callback():
-    # del st.session_state.product
-    # del st.session_state.gender
-    # del st.session_state.profession
-    # del st.session_state.hobby
-#     placeholder.empty()
-
-# import streamlit as st
-# import time
-
-# class SessionState:
-#     def __init__(self):
-#         self.product = ""
-#         self.gender = ""
-#         self.profession = ""
-#         self.hobby = ""
-
-# state = SessionState()
-
-# def submit_callback():
-#     with output_placeholder.container():
-#         st.write("Product: " + state.product)
-#         st.write("Gender: " + state.gender) 
-#         st.write("Profession: " + state.profession) 
-#         st.write("Hobby: "+ state.hobby)
-
-# def delete_callback():
-#     state.product = ""
-#     state.gender = ""
-#     state.profession = ""
-#     state.hobby = ""
-#     output_placeholder.empty()
-
-# form_placeholder = st.empty()
-# output_placeholder = st.empty()
-
-# submitted = False
-# clear = False
-
-# with form_placeholder.form("my_form", clear_on_submit=True):
-#     st.write("Input")
-#     # product
-#     state.product = st.text_input("product", value=state.product, key="product")
-#     # gender
-#     state.gender = st.radio("gender", ["male", "female"], value=state.gender, key="gender")
-#     # profession
-#     state.profession = st.text_input("profession", value=state.profession, key="profession")
-#     # hobby
-#     state.hobby = st.text_input("hobby", value=state.hobby, key="hobby")
-#     # Every form must have a submit button.
-#     btn1, btn2=st.columns(2)
-#     with btn1:
-#         submitted = st.form_submit_button(label='Submit')
-#     with btn2:
-#         clear = st.form_submit_button(label='Clear')
-
-# if clear:
-#     delete_callback()
-# if submitted:
-#     submit_callback()
-
-
-# def submit_callback():
-#     with output_placeholder.container():
-#         st.write("Product: " + st.session_state.product)
-#         st.write("Gender: " + st.session_state.gender) 
-#         st.write("Profession: " + st.session_state.profession) 
-#         st.write("Hobby: "+ st.session_state.hobby)
-
-# def delete_callback():
-#     del st.session_state.product
-#     del st.session_state.gender
-#     del st.session_state.profession
-#     del st.session_state.hobby
-#     output_placeholder.empty()
-
-# form_placeholder = st.empty()
-# output_placeholder = st.empty()
-
-# with form_placeholder.form("my_input", clear_on_submit=True):
-#     st.write("Input")
-#     # product
-#     product=st.text_input("product", key="product")
-#     # gender
-#     gender=st.radio("gender", ["male", "female"], key="gender")
-#     # profession
-#     profession=st.text_input("profession", key="profession")
-#     # hobby
-#     hobby=st.text_input("hobby", key="hobby")
-#     # Every form must have a submit button.
-#     btn1, btn2=st.columns(2)
-#     with btn1:
-#             submitted = st.form_submit_button(label='Submit', on_click=submit_callback)
-#             # submitted = st.form_submit_button(label='Submit')
-#     with btn2:
-#             clear = st.form_submit_button(label='Clear', on_click=delete_callback) 
-    # if submitted:
-    #     # handle the user_input here
-    #     st.write("Product: " + st.session_state.product)
-    #     st.write("Gender: " + st.session_state.gender) 
-    #     st.write("Profession: " + st.session_state.profession) 
-    #     st.write("Hobby: "+ st.session_state.hobby)
-        
-    # if clear:
-    #     del st.session_state.product
-    #     del st.session_state.gender
-    #     del st.session_state.profession
-    #     del st.session_state.hobby
-    #     placeholder.empty()
-    
-
-
-
-# # with col2:
-# _LOREM_IPSUM = (
-#     "product" + str(st.session_state.product) + "\n"
-#     + "gender" + str(st.session_state.gender) + "\n"
-#     + "profession" + str(st.session_state.profession) + "\n"
-#     + "hobby" + str(st.session_state.hobby) +"\n"
-# )
-
-
-# def stream_data():
-#     for word in _LOREM_IPSUM.split(" "):
-#         yield word + " "
-#         time.sleep(0.02)
-
-# if st.button("Stream data"):
-#     st.write_stream(stream_data)
-# placeholder = st.empty()
-
-# with placeholder.container():
-#     st.write("Hello")
-
-
-# import streamlit as st
-
-# class SessionState:
-#     def __init__(self):
-#         self.input_text = ""
-
-# state = SessionState()
-
-# with st.form("my_form"):
-#     state.input_text = st.text_input("Enter something:", value=state.input_text)
-#     clear_button = st.form_submit_button("Clear")
-#     submit_button = st.form_submit_button("Submit")
-
-#     if clear_button:
-#         state.input_text = ""
-#     if submit_button:
-#         # handle the user_input here
-#         st.write(f"You submitted: {state.input_text}")
-
-# import streamlit as st
-# import time
-
-# class SessionState:
-#     def __init__(self):
-#         self.product = ""
-#         self.gender = ""
-#         self.profession = ""
-#         self.hobby = ""
-
-# state = SessionState()
-
-# def submit_callback():
-#     output_placeholder.write("Product: " + state.product)
-#     output_placeholder.write("Gender: " + state.gender) 
-#     output_placeholder.write("Profession: " + state.profession) 
-#     output_placeholder.write("Hobby: "+ state.hobby)
-
-# def delete_callback():
-#     state.product = ""
-#     state.gender = ""
-#     state.profession = ""
-#     state.hobby = ""
-#     output_placeholder.empty()
-
-# form_placeholder = st.empty()
-# output_placeholder = st.empty()
-
-# with form_placeholder.form("my_form"):
-#     st.write("Input")
-#     # product
-#     state.product = st.text_input("product", value=state.product, key="product")
-#     # gender
-#     state.gender = st.radio("gender", ["male", "female"], value=state.gender, key="gender")
-#     # profession
-#     state.profession = st.text_input("profession", value=state.profession, key="profession")
-#     # hobby
-#     state.hobby = st.text_input("hobby", value=state.hobby, key="hobby")
-#     # Every form must have a submit button.
-#     btn1, btn2=st.columns(2)
-#     with btn1:
-#         submitted = st.form_submit_button(label='Submit')
-#     with btn2:
-#         clear = st.form_submit_button(label='Clear') 
-
-#     if clear:
-#         delete_callback()
-#     if submitted:
-#         # handle the user_input here
-#         submit_callback()
-
-    
-
-# _LOREM_IPSUM = """
-# Lorem ipsum dolor sit amet, **consectetur adipiscing** elit, sed do eiusmod tempor
-# incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis
-# nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat.
-# """
-
-# def stream_data():
-#     for word in _LOREM_IPSUM.split(" "):
-#         yield word + " "
-#         time.sleep(0.02)
-
-#     for word in _LOREM_IPSUM.split(" "):
-#         yield word + " "
-#         time.sleep(0.02)
-
-
-# if st.button("Stream data"):
-#     st.write_stream(stream_data)
-
-
-# import streamlit as st
-# import SessionState
-# from streamlit.server.server import Server
-# import streamlit.report_thread as ReportThread
-
-# # Initialize session state
-# session_state = SessionState.get(product="", gender="", profession="", hobby="")
-
-# # Create a form for user inputs
-# with st.form("my_input"):
-#     st.write("Input")
-#     # product
-#     product = st.text_input("product", value=session_state.product)
-#     # gender
-#     gender = st.radio("gender", ["male", "female"], index=["male", "female"].index(session_state.gender) if session_state.gender else 0)
-#     # profession
-#     profession = st.text_input("profession", value=session_state.profession)
-#     # hobby
-#     hobby = st.text_input("hobby", value=session_state.hobby)
-
-#     # Every form must have a submit button.
-#     submitted = st.form_submit_button("Submit")
-#     clear = st.form_submit_button("Clear")
-
-# # Display the user inputs
-# with st.form("my_output"):
-#     if submitted and not clear:
-#         # Save inputs to session state
-#         session_state.product = product
-#         session_state.gender = gender
-#         session_state.profession = profession
-#         session_state.hobby = hobby
-
-#         st.write("product", product)
-#         st.write("gender", gender)
-#         st.write("profession", profession)
-#         st.write("hobby", hobby)
-
-#     # Clear the user inputs
-#     if clear:
-#         # Clear session state
-#         session_state.product = ""
-#         session_state.gender = ""
-#         session_state.profession = ""
-#         session_state.hobby = ""
-
-#         st.experimental_rerun()

utils.py

@@ -0,0 +1,40 @@
+from transformers import pipeline, AutoModelForCausalLM, AutoTokenizer
+import re
+
+# Load saved model and tokenizer
+model_checkpoint = "24NLPGroupO/EmailGeneration"
+tokenizer = AutoTokenizer.from_pretrained(model_checkpoint, truncation=True)
+model = AutoModelForCausalLM.from_pretrained(model_checkpoint)
+
+# Set up the generation pipeline
+generator = pipeline('text-generation', model=model, tokenizer=tokenizer)
+
+def clean_generated_text(text):
+    # Basic cleaning
+    text = re.sub(r'^(Re:|Fwd:)', '', text)  # Remove reply and forward marks
+    text = re.sub(r'Best regards,.*$', '', text, flags=re.DOTALL)  # Remove everything after signature
+    text = re.sub(r'PHONE.*$', '', text, flags=re.DOTALL)  # Remove everything after phone numbers
+    text = re.sub(r'Email:.*$', '', text, flags=re.DOTALL)  # Remove everything after email addresses
+    text = re.sub(r'Cc:.*$', '', text, flags=re.DOTALL)  # Remove CC list
+    text = re.sub(r'\* Attachments:.*', '', text, flags=re.S)  # Remove 'Attachments:' and everything following it
+    text = re.sub(r'©️ .*$', '', text, flags=re.DOTALL)  # Remove copyright and ownership statements
+    text = re.sub(r'URL If this message is not displaying properly, click here.*$', '', text, flags=re.DOTALL)  # Remove error display message and links
+    text = re.sub(r'\d{5,}', 'NUMBER', text)  # Replace long sequences of numbers, likely phone numbers or ZIP codes
+    return text.strip()
+
+def generate_email(product, gender, profession, hobby):
+    input_text = f"{product} {gender} {profession} {hobby}"
+    result = generator(
+        input_text,                 # Initial text to prompt the model. Sets the context or topic for text generation.
+        max_length=256,             # Maximum length of the generated text in tokens, limiting the output size.
+        do_sample=True,             # Enables stochastic sampling; the model can generate diverse outputs at each step.
+        top_k=20,                   # Limits the vocabulary considered at each step to the top-k most likely next words.
+        top_p=0.6,                  # Uses nucleus sampling: Narrows down to the smallest set of words totaling 60% of the likelihood.
+        temperature=0.4,            # Scales logits before sampling to reduce randomness and produce more deterministic output.
+        repetition_penalty=1.5,     # Penalizes words that were already mentioned, reducing repetition in the text.
+        # truncation=True,            # Truncates the output to the maximum length if it exceeds it.
+        num_return_sequences=3      # Generates three different sequences to choose from, enhancing output variety.
+        ) 
+    # Select the best output from the generated sequences
+    best_text = sorted([clean_generated_text(r['generated_text']) for r in result], key=len)[-1]
+    return best_text