Update app.py

app.py

@@ -1,442 +1,442 @@
-import pandas as pd
-from PIL import Image
-import streamlit as st
-import cv2
-from streamlit_drawable_canvas import st_canvas
-import torch
-from diffusers import AutoPipelineForInpainting
-import numpy as np
-from streamlit_image_select import image_select
-import os
-import requests 
-from streamlit_navigation_bar import st_navbar
-from langchain_community.llms import Ollama
-import base64
-from io import BytesIO
-from PIL import Image, ImageDraw
-from streamlit_lottie import st_lottie 
-from streamlit_option_menu import option_menu
-import json
-from transformers import pipeline
-import streamlit as st
-from streamlit_modal import Modal
-import streamlit.components.v1 as components
-from datetime import datetime
-
-
-def image_to_base64(image_path):
-    with open(image_path, "rb") as img_file:
-        return base64.b64encode(img_file.read()).decode()
-    
-
-@st.cache_resource
-def load_model():
-    pipeline_ = AutoPipelineForInpainting.from_pretrained("kandinsky-community/kandinsky-2-2-decoder-inpaint", torch_dtype=torch.float16).to("cuda")
-    return pipeline_
-
-# @st.cache_resource
-def prompt_improvment(pre_prompt):
-
-    llm = Ollama(model="llama3:latest",num_ctx=1000)
-    enhancement="Please use details from the prompt mentioned above, focusing only what user is thinking with the prompt and also add 8k resolution. Its a request only provide image description and brief prompt no other text."
-    prompt = pre_prompt+"\n"+enhancement
-    # result = llm.invoke(prompt)
-    return llm.stream(prompt)
-def numpy_to_list(array):
-
-    current=[]
-    for value in array:
-        if isinstance(value,type(np.array([]))):
-            result=numpy_to_list(value)
-            current.append(result)
-        else:
-            
-            current.append(int(value))
-    return current
-
-
-
-@st.cache_resource
-def llm_text_response():
-    llm = Ollama(model="llama3:latest",num_ctx=1000)
-    return llm.stream
-
-def model_single_out(prompt):
-    pipe=load_model()
-    image = pipe(prompt).images[0]
-    return image
-
-def model_out_put(init_image,mask_image,prompt,negative_prompt):
-    pipeline_ = load_model()
-    image = pipeline_(prompt=prompt, negative_prompt=negative_prompt, image=init_image, mask_image=mask_image).images[0]
-    return image
-
-@st.cache_resource
-def multimodel():
-    pipeline_ = pipeline("text-classification", model = "model_collection\model_4")
-    return pipeline_
-   
-def multimodel_output(prompt):
-    pipeline_ = multimodel()
-    image = pipeline_(prompt)
-    return image[0]['label']
-
-def d4_to_3d(image):
-    formatted_array=[]
-    for j in image:
-        neste_list=[]
-        for k in j:
-            if any([True if i>0 else False for i in k]):
-                neste_list.append(True)
-            else:
-                neste_list.append(False)
-        formatted_array.append(neste_list)
-    print(np.shape(formatted_array))
-    return np.array(formatted_array)
-    
-st.set_page_config(layout="wide")
-
-st.write(str(os.getcwd()))
-
-img_selection=None
-# Specify canvas parameters in application
-drawing_mode = st.sidebar.selectbox(
-    "Drawing tool:", ("freedraw","point", "line", "rect", "circle", "transform")
-)
-
-
-dictionary=st.session_state
-if "every_prompt_with_val" not in dictionary:
-    dictionary['every_prompt_with_val']=[]
-if "current_image" not in dictionary:
-    dictionary['current_image']=[]
-if "prompt_collection"  not in dictionary:
-    dictionary['prompt_collection']=[]
-if "user" not in dictionary:
-    dictionary['user']=None
-if "current_session" not in dictionary:
-    dictionary['current_session']=None
-
-stroke_width = st.sidebar.slider("Stroke width: ", 1, 25, 20)
-if drawing_mode == 'point':
-        point_display_radius = st.sidebar.slider("Point display radius: ", 1, 25, 3)
-stroke_color = '#000000'
-bg_color = "#eee"
-
-
-column1,column2=st.columns([0.7,0.35])
-
-with open("DataBase\datetimeRecords.json","r") as read:
-    dateTimeRecord=json.load(read)
-with column2:
-    st.header("HISTORY")
-    tab1,tab2,tab3,tab4=st.tabs(["CHAT HISTORY","IMAGES","PROMPT IMPROVEMENT","LOGIN"])
-    with tab1:
-
-        
- 
-        if not len(dictionary['every_prompt_with_val']):
-            st.header("I will store all the chat for the current session")
-            with open("lotte_animation_saver\\animation_4.json") as read:
-                url_json=json.load(read)
-            st_lottie(url_json,height = 400)
-        else:
-
-            with st.container(height=600):
-
-
-                for index,prompts_ in enumerate(dictionary['every_prompt_with_val'][::-1]):
-                    if prompts_[-1]=="@working":
-                        if index==0:
-                            st.write(prompts_[0].upper())
-                            data_need=st.write_stream(llm_text_response()(prompts_[0]))
-                            dictionary['every_prompt_with_val'][-1]=(prompts_[0],str(data_need))
-                            
-                    elif isinstance(prompts_[-1],str):
-                        if index==0:
-                            st.text_area(label=prompts_[0].upper(),value=prompts_[-1],height=500)
-                        else:
-                            st.text_area(label=prompts_[0].upper(),value=prompts_[-1])
-
-                    else:
-                        st.write(prompts_[0].upper())
-                        with st.container(height=400):
-                            format1,format2=st.columns([0.2,0.8])
-                            with format1:
-                                new_img=Image.open("ALL_image_formation\image_gen.png")
-                                st.write("<br>",unsafe_allow_html=True)
-                                size = min(new_img.size)
-                                mask = Image.new('L', (size, size), 0)
-                                draw = ImageDraw.Draw(mask)
-                                draw.ellipse((0, 0, size, size), fill=255)
-
-                                image = new_img.crop((0, 0, size, size))
-                                image.putalpha(mask)
-                                st.image(image)                    
-                            with format2:
-
-                                st.write("<br>",unsafe_allow_html=True)
-                                size = min(prompts_[-1].size)
-                                mask = Image.new('L', (size, size), 0)
-                                draw = ImageDraw.Draw(mask)
-                                draw.ellipse((0, 0, size, size), fill=255)
-
-                                # Crop the image to a square and apply the mask
-                                image = prompts_[-1].crop((0, 0, size, size))
-                                image.putalpha(mask)
-                                st.image(image)
- 
-    with tab2:
-        
-        if "current_image" in dictionary and len(dictionary['current_image']):
-            with st.container(height=600):
-                dictinory_length=len(dictionary['current_image'])
-                
-                img_selection = image_select(
-                    label="",
-                    images=dictionary['current_image'] if len(dictionary['current_image'])!=0 else None,
-                )
-                if img_selection in dictionary['current_image']:
-                    dictionary['current_image'].remove(img_selection)
-                    dictionary['current_image'].insert(0,img_selection)
-                    # st.rerun()
-
-                img_selection.save("image.png")
-                with open("image.png", "rb") as file:
-                    downl=st.download_button(label="DOWNLOAD",data=file,file_name="image.png",mime="image/png")
-                os.remove("image.png")
-        else:
-
-            st.header("This section will store the updated images")
-            with open("lotte_animation_saver\\animation_1.json") as read:
-                url_json=json.load(read)
-            st_lottie(url_json,height = 400)
-    with tab3:
-        if len(dictionary['prompt_collection'])!=0:
-            with st.container(height=600):
-                prompt_selection=st.selectbox(label="Select the prompt for improvment",options=["Mention below are prompt history"]+dictionary["prompt_collection"],index=0)
-
-                if prompt_selection!="Mention below are prompt history":
-
-                    generated_prompt=prompt_improvment(prompt_selection)
-                    dictionary['generated_image_prompt'].append(generated_prompt)
-                    st.write_stream(generated_prompt)
-
-        else:
-
-            st.header("This section will provide prompt improvement section")
-            with open("lotte_animation_saver\\animation_3.json") as read:
-                url_json=json.load(read)
-            st_lottie(url_json,height = 400)
-        with tab4:
-            
-            # with st.container(height=600):
-
-            if not dictionary['user']   : 
-                with st.form("my_form"):
-                    # st.header("Please login for save your data")
-                    with open("lotte_animation_saver\\animation_5.json") as read:
-                        url_json=json.load(read)
-                    st_lottie(url_json,height = 200)                
-                    user_id = st.text_input("user login")
-                    password = st.text_input("password",type="password")
-                    submitted_login = st.form_submit_button("Submit")
-                    # Every form must have a submit button.
-
-                    if submitted_login:
-                        with open("DataBase\login.json","r") as read:
-                            login_base=json.load(read)
-                        if user_id in login_base and login_base[user_id]==password:
-                            dictionary['user']=user_id
-                            st.rerun()
-                        else:
-                            st.error("userid or password incorrect")
-
-                        st.write("working")
-                    modal = Modal(
-                        "Sign up", 
-                        key="demo-modal",
-                        
-                        padding=10,    # default value
-                        max_width=600  # default value
-                    )
-                open_modal = st.button("sign up")
-                if open_modal:
-                    modal.open()
-
-                if modal.is_open():
-                    with modal.container():
-
-                        with st.form("my_form1"):
-                            sign_up_column_left,sign_up_column_right=st.columns(2)
-                            with sign_up_column_left:
-                                with open("lotte_animation_saver\\animation_6.json") as read:
-                                    url_json=json.load(read)
-                                st_lottie(url_json,height = 200) 
-    
-                            with sign_up_column_right:
-                                user_id = st.text_input("user login")
-                                password = st.text_input("password",type="password")
-                                submitted_signup = st.form_submit_button("Submit")
-
-                            if submitted_signup:
-                                with open("DataBase\login.json","r") as read:
-                                    login_base=json.load(read)
-                                if not login_base:
-                                    login_base={}
-                                if user_id not in login_base:
-                                    login_base[user_id]=password
-                                    with open("DataBase\login.json","w") as write:
-                                        json.dump(login_base,write,indent=2)  
-                                    st.success("you are a part now")  
-                                    dictionary['user']=user_id
-                                    modal.close()                       
-                                else:
-                                    st.error("user id already exists")
-            else:
-                st.header("REPORTED ISSUES")
-                with st.container(height=370):
-
-                    with open("DataBase\datetimeRecords.json") as feedback:
-                        temp_issue=json.load(feedback)
-
-                    arranged_feedback=reversed(temp_issue['database'])
-                    
-                    for report in arranged_feedback:
-                        user_columns,user_feedback=st.columns([0.3,0.8])
-
-                        with user_columns:
-                            st.write(report[-1])
-                        with user_feedback:
-                            st.write(report[1])
-                     
-                feedback=st.text_area("Feedback Report and Improvement",placeholder="")
-                summit=st.button("submit")
-                if summit:
-                    with open("DataBase\datetimeRecords.json","r") as feedback_sumit:
-                        temp_issue_submit=json.load(feedback_sumit)        
-                    if  "database" not in temp_issue_submit:
-                        temp_issue_submit["database"]=[]
-                    temp_issue_submit["database"].append((str(datetime.now()),feedback,dictionary['user'])) 
-                    with open("DataBase\datetimeRecords.json","w") as feedback_sumit:
-                        json.dump(temp_issue_submit,feedback_sumit)                    
-                            
-
-
-                    # st.rerun()
-                
-                
-                    
-
-
-
-bg_image = st.sidebar.file_uploader("PLEASE UPLOAD IMAGE FOR EDITING:", type=["png", "jpg"])
-bg_doc = st.sidebar.file_uploader("PLEASE UPLOAD DOC FOR PPT/PDF/STORY:", type=["pdf","xlsx"])
-
-
-if "bg_image" not in dictionary:
-    dictionary["bg_image"]=None
-
-if img_selection  and dictionary['bg_image']==bg_image:
-    gen_image=dictionary['current_image'][0]
-else:
-    if bg_image:
-        gen_image=Image.open(bg_image) 
-    else:
-        gen_image=None
-
-
-
-
-
-
-with column1:
-# Create a canvas component
-    changes,implementation,current=st.columns([0.3,0.6,0.3])
-
-    with implementation:
-                st.write("<br>"*5,unsafe_allow_html=True)
-                canvas_result = st_canvas(
-                    fill_color="rgba(255, 165, 0, 0.3)",  # Fixed fill color with some opacity
-                    stroke_width=stroke_width,
-                    stroke_color=stroke_color,
-                    background_color=bg_color,
-                    background_image=gen_image if gen_image else Image.open("ALL_image_formation\image_gen.png"),
-                    update_streamlit=True,
-                    height=500,
-                    width=500,
-                    drawing_mode=drawing_mode,
-                    point_display_radius=point_display_radius if drawing_mode == 'point' else 0,
-                    key="canvas",
-                )
-
-
-
-
-
-with column1:
-    # prompt=st.text_area("Please provide the prompt")
-    prompt=st.chat_input("Please provide the prompt")
-    
-    negative_prompt="the black masked area"
-
-    # run=st.button("run_experiment")
-
-
-
-if canvas_result.image_data is not None:
-    if prompt:
-
-        text_or_image=multimodel_output(prompt)
-        
-        if text_or_image=="LABEL_0":
-        
-            if "generated_image_prompt" not in dictionary:
-                dictionary['generated_image_prompt']=[]
-            if prompt not in dictionary['prompt_collection'] and prompt not in dictionary['generated_image_prompt']:
-                dictionary['prompt_collection']=[prompt]+dictionary['prompt_collection']
-            new_size=np.array(canvas_result.image_data).shape[:2]
-            new_size=(new_size[-1],new_size[0])
-            if bg_image!=dictionary["bg_image"] :
-                dictionary["bg_image"]=bg_image
-                if bg_image!=None:
-                    imf=Image.open(bg_image).resize(new_size)
-                else:
-                    with open("lotte_animation_saver/animation_4.json") as read:
-                        url_json=json.load(read)        
-                    st_lottie(url_json) 
-                    imf=Image.open("ALL_image_formation\home_screen.jpg").resize(new_size)
-            else:
-                if len(dictionary['current_image'])!=0:
-                    imf=dictionary['current_image'][0]
-                else:
-                    with open("lotte_animation_saver/animation_4.json") as read:
-                        url_json=json.load(read)        
-                    st_lottie(url_json) 
-                    imf=Image.open("ALL_image_formation\home_screen.jpg")
-
-            negative_image =d4_to_3d(np.array(canvas_result.image_data))
-            if np.sum(negative_image)==0:
-                negative_image=Image.fromarray(np.where(negative_image == False, True, negative_image))
-            else:
-                negative_image=Image.fromarray(negative_image)
-            
-            modifiedValue=model_out_put(imf,negative_image,prompt,negative_prompt)
-            modifiedValue.save("ALL_image_formation/current_session_image.png")
-            dictionary['current_image']=[modifiedValue]+dictionary['current_image']
-            dictionary['every_prompt_with_val'].append((prompt,modifiedValue))
-            st.rerun()
-        else:
-            st.write("nothing importent")
-            modifiedValue="@working"
-            dictionary['every_prompt_with_val'].append((prompt,modifiedValue))
-            st.rerun()
-            # st.image(modifiedValue,width=300)
-    
-        
-        
-if canvas_result.json_data is not None:
-    objects = pd.json_normalize(canvas_result.json_data["objects"]) # need to convert obj to str because PyArrow
-    for col in objects.select_dtypes(include=['object']).columns:
-        objects[col] = objects[col].astype("str")
-
+import pandas as pd
+from PIL import Image
+import streamlit as st
+import cv2
+from streamlit_drawable_canvas import st_canvas
+import torch
+from diffusers import AutoPipelineForInpainting
+import numpy as np
+from streamlit_image_select import image_select
+import os
+import requests 
+from streamlit_navigation_bar import st_navbar
+from langchain_community.llms import Ollama
+import base64
+from io import BytesIO
+from PIL import Image, ImageDraw
+from streamlit_lottie import st_lottie 
+from streamlit_option_menu import option_menu
+import json
+from transformers import pipeline
+import streamlit as st
+from streamlit_modal import Modal
+import streamlit.components.v1 as components
+from datetime import datetime
+
+
+def image_to_base64(image_path):
+    with open(image_path, "rb") as img_file:
+        return base64.b64encode(img_file.read()).decode()
+    
+
+@st.cache_resource
+def load_model():
+    pipeline_ = AutoPipelineForInpainting.from_pretrained("kandinsky-community/kandinsky-2-2-decoder-inpaint", torch_dtype=torch.float16).to("cuda")
+    return pipeline_
+
+# @st.cache_resource
+def prompt_improvment(pre_prompt):
+
+    llm = Ollama(model="llama3:latest",num_ctx=1000)
+    enhancement="Please use details from the prompt mentioned above, focusing only what user is thinking with the prompt and also add 8k resolution. Its a request only provide image description and brief prompt no other text."
+    prompt = pre_prompt+"\n"+enhancement
+    # result = llm.invoke(prompt)
+    return llm.stream(prompt)
+def numpy_to_list(array):
+
+    current=[]
+    for value in array:
+        if isinstance(value,type(np.array([]))):
+            result=numpy_to_list(value)
+            current.append(result)
+        else:
+            
+            current.append(int(value))
+    return current
+
+
+
+@st.cache_resource
+def llm_text_response():
+    llm = Ollama(model="llama3:latest",num_ctx=1000)
+    return llm.stream
+
+def model_single_out(prompt):
+    pipe=load_model()
+    image = pipe(prompt).images[0]
+    return image
+
+def model_out_put(init_image,mask_image,prompt,negative_prompt):
+    pipeline_ = load_model()
+    image = pipeline_(prompt=prompt, negative_prompt=negative_prompt, image=init_image, mask_image=mask_image).images[0]
+    return image
+
+@st.cache_resource
+def multimodel():
+    pipeline_ = pipeline("text-classification", model = "model_collection\model_4")
+    return pipeline_
+   
+def multimodel_output(prompt):
+    pipeline_ = multimodel()
+    image = pipeline_(prompt)
+    return image[0]['label']
+
+def d4_to_3d(image):
+    formatted_array=[]
+    for j in image:
+        neste_list=[]
+        for k in j:
+            if any([True if i>0 else False for i in k]):
+                neste_list.append(True)
+            else:
+                neste_list.append(False)
+        formatted_array.append(neste_list)
+    print(np.shape(formatted_array))
+    return np.array(formatted_array)
+    
+st.set_page_config(layout="wide")
+
+st.write(str(os.getcwd()))
+
+img_selection=None
+# Specify canvas parameters in application
+drawing_mode = st.sidebar.selectbox(
+    "Drawing tool:", ("freedraw","point", "line", "rect", "circle", "transform")
+)
+
+
+dictionary=st.session_state
+if "every_prompt_with_val" not in dictionary:
+    dictionary['every_prompt_with_val']=[]
+if "current_image" not in dictionary:
+    dictionary['current_image']=[]
+if "prompt_collection"  not in dictionary:
+    dictionary['prompt_collection']=[]
+if "user" not in dictionary:
+    dictionary['user']=None
+if "current_session" not in dictionary:
+    dictionary['current_session']=None
+
+stroke_width = st.sidebar.slider("Stroke width: ", 1, 25, 20)
+if drawing_mode == 'point':
+        point_display_radius = st.sidebar.slider("Point display radius: ", 1, 25, 3)
+stroke_color = '#000000'
+bg_color = "#eee"
+
+
+column1,column2=st.columns([0.7,0.35])
+
+with open("\home\user\DataBase\datetimeRecords.json","r") as read:
+    dateTimeRecord=json.load(read)
+with column2:
+    st.header("HISTORY")
+    tab1,tab2,tab3,tab4=st.tabs(["CHAT HISTORY","IMAGES","PROMPT IMPROVEMENT","LOGIN"])
+    with tab1:
+
+        
+ 
+        if not len(dictionary['every_prompt_with_val']):
+            st.header("I will store all the chat for the current session")
+            with open("lotte_animation_saver\\animation_4.json") as read:
+                url_json=json.load(read)
+            st_lottie(url_json,height = 400)
+        else:
+
+            with st.container(height=600):
+
+
+                for index,prompts_ in enumerate(dictionary['every_prompt_with_val'][::-1]):
+                    if prompts_[-1]=="@working":
+                        if index==0:
+                            st.write(prompts_[0].upper())
+                            data_need=st.write_stream(llm_text_response()(prompts_[0]))
+                            dictionary['every_prompt_with_val'][-1]=(prompts_[0],str(data_need))
+                            
+                    elif isinstance(prompts_[-1],str):
+                        if index==0:
+                            st.text_area(label=prompts_[0].upper(),value=prompts_[-1],height=500)
+                        else:
+                            st.text_area(label=prompts_[0].upper(),value=prompts_[-1])
+
+                    else:
+                        st.write(prompts_[0].upper())
+                        with st.container(height=400):
+                            format1,format2=st.columns([0.2,0.8])
+                            with format1:
+                                new_img=Image.open("ALL_image_formation\image_gen.png")
+                                st.write("<br>",unsafe_allow_html=True)
+                                size = min(new_img.size)
+                                mask = Image.new('L', (size, size), 0)
+                                draw = ImageDraw.Draw(mask)
+                                draw.ellipse((0, 0, size, size), fill=255)
+
+                                image = new_img.crop((0, 0, size, size))
+                                image.putalpha(mask)
+                                st.image(image)                    
+                            with format2:
+
+                                st.write("<br>",unsafe_allow_html=True)
+                                size = min(prompts_[-1].size)
+                                mask = Image.new('L', (size, size), 0)
+                                draw = ImageDraw.Draw(mask)
+                                draw.ellipse((0, 0, size, size), fill=255)
+
+                                # Crop the image to a square and apply the mask
+                                image = prompts_[-1].crop((0, 0, size, size))
+                                image.putalpha(mask)
+                                st.image(image)
+ 
+    with tab2:
+        
+        if "current_image" in dictionary and len(dictionary['current_image']):
+            with st.container(height=600):
+                dictinory_length=len(dictionary['current_image'])
+                
+                img_selection = image_select(
+                    label="",
+                    images=dictionary['current_image'] if len(dictionary['current_image'])!=0 else None,
+                )
+                if img_selection in dictionary['current_image']:
+                    dictionary['current_image'].remove(img_selection)
+                    dictionary['current_image'].insert(0,img_selection)
+                    # st.rerun()
+
+                img_selection.save("image.png")
+                with open("image.png", "rb") as file:
+                    downl=st.download_button(label="DOWNLOAD",data=file,file_name="image.png",mime="image/png")
+                os.remove("image.png")
+        else:
+
+            st.header("This section will store the updated images")
+            with open("lotte_animation_saver\\animation_1.json") as read:
+                url_json=json.load(read)
+            st_lottie(url_json,height = 400)
+    with tab3:
+        if len(dictionary['prompt_collection'])!=0:
+            with st.container(height=600):
+                prompt_selection=st.selectbox(label="Select the prompt for improvment",options=["Mention below are prompt history"]+dictionary["prompt_collection"],index=0)
+
+                if prompt_selection!="Mention below are prompt history":
+
+                    generated_prompt=prompt_improvment(prompt_selection)
+                    dictionary['generated_image_prompt'].append(generated_prompt)
+                    st.write_stream(generated_prompt)
+
+        else:
+
+            st.header("This section will provide prompt improvement section")
+            with open("lotte_animation_saver\\animation_3.json") as read:
+                url_json=json.load(read)
+            st_lottie(url_json,height = 400)
+        with tab4:
+            
+            # with st.container(height=600):
+
+            if not dictionary['user']   : 
+                with st.form("my_form"):
+                    # st.header("Please login for save your data")
+                    with open("lotte_animation_saver\\animation_5.json") as read:
+                        url_json=json.load(read)
+                    st_lottie(url_json,height = 200)                
+                    user_id = st.text_input("user login")
+                    password = st.text_input("password",type="password")
+                    submitted_login = st.form_submit_button("Submit")
+                    # Every form must have a submit button.
+
+                    if submitted_login:
+                        with open("DataBase\login.json","r") as read:
+                            login_base=json.load(read)
+                        if user_id in login_base and login_base[user_id]==password:
+                            dictionary['user']=user_id
+                            st.rerun()
+                        else:
+                            st.error("userid or password incorrect")
+
+                        st.write("working")
+                    modal = Modal(
+                        "Sign up", 
+                        key="demo-modal",
+                        
+                        padding=10,    # default value
+                        max_width=600  # default value
+                    )
+                open_modal = st.button("sign up")
+                if open_modal:
+                    modal.open()
+
+                if modal.is_open():
+                    with modal.container():
+
+                        with st.form("my_form1"):
+                            sign_up_column_left,sign_up_column_right=st.columns(2)
+                            with sign_up_column_left:
+                                with open("lotte_animation_saver\\animation_6.json") as read:
+                                    url_json=json.load(read)
+                                st_lottie(url_json,height = 200) 
+    
+                            with sign_up_column_right:
+                                user_id = st.text_input("user login")
+                                password = st.text_input("password",type="password")
+                                submitted_signup = st.form_submit_button("Submit")
+
+                            if submitted_signup:
+                                with open("DataBase\login.json","r") as read:
+                                    login_base=json.load(read)
+                                if not login_base:
+                                    login_base={}
+                                if user_id not in login_base:
+                                    login_base[user_id]=password
+                                    with open("DataBase\login.json","w") as write:
+                                        json.dump(login_base,write,indent=2)  
+                                    st.success("you are a part now")  
+                                    dictionary['user']=user_id
+                                    modal.close()                       
+                                else:
+                                    st.error("user id already exists")
+            else:
+                st.header("REPORTED ISSUES")
+                with st.container(height=370):
+
+                    with open("DataBase\datetimeRecords.json") as feedback:
+                        temp_issue=json.load(feedback)
+
+                    arranged_feedback=reversed(temp_issue['database'])
+                    
+                    for report in arranged_feedback:
+                        user_columns,user_feedback=st.columns([0.3,0.8])
+
+                        with user_columns:
+                            st.write(report[-1])
+                        with user_feedback:
+                            st.write(report[1])
+                     
+                feedback=st.text_area("Feedback Report and Improvement",placeholder="")
+                summit=st.button("submit")
+                if summit:
+                    with open("DataBase\datetimeRecords.json","r") as feedback_sumit:
+                        temp_issue_submit=json.load(feedback_sumit)        
+                    if  "database" not in temp_issue_submit:
+                        temp_issue_submit["database"]=[]
+                    temp_issue_submit["database"].append((str(datetime.now()),feedback,dictionary['user'])) 
+                    with open("DataBase\datetimeRecords.json","w") as feedback_sumit:
+                        json.dump(temp_issue_submit,feedback_sumit)                    
+                            
+
+
+                    # st.rerun()
+                
+                
+                    
+
+
+
+bg_image = st.sidebar.file_uploader("PLEASE UPLOAD IMAGE FOR EDITING:", type=["png", "jpg"])
+bg_doc = st.sidebar.file_uploader("PLEASE UPLOAD DOC FOR PPT/PDF/STORY:", type=["pdf","xlsx"])
+
+
+if "bg_image" not in dictionary:
+    dictionary["bg_image"]=None
+
+if img_selection  and dictionary['bg_image']==bg_image:
+    gen_image=dictionary['current_image'][0]
+else:
+    if bg_image:
+        gen_image=Image.open(bg_image) 
+    else:
+        gen_image=None
+
+
+
+
+
+
+with column1:
+# Create a canvas component
+    changes,implementation,current=st.columns([0.3,0.6,0.3])
+
+    with implementation:
+                st.write("<br>"*5,unsafe_allow_html=True)
+                canvas_result = st_canvas(
+                    fill_color="rgba(255, 165, 0, 0.3)",  # Fixed fill color with some opacity
+                    stroke_width=stroke_width,
+                    stroke_color=stroke_color,
+                    background_color=bg_color,
+                    background_image=gen_image if gen_image else Image.open("ALL_image_formation\image_gen.png"),
+                    update_streamlit=True,
+                    height=500,
+                    width=500,
+                    drawing_mode=drawing_mode,
+                    point_display_radius=point_display_radius if drawing_mode == 'point' else 0,
+                    key="canvas",
+                )
+
+
+
+
+
+with column1:
+    # prompt=st.text_area("Please provide the prompt")
+    prompt=st.chat_input("Please provide the prompt")
+    
+    negative_prompt="the black masked area"
+
+    # run=st.button("run_experiment")
+
+
+
+if canvas_result.image_data is not None:
+    if prompt:
+
+        text_or_image=multimodel_output(prompt)
+        
+        if text_or_image=="LABEL_0":
+        
+            if "generated_image_prompt" not in dictionary:
+                dictionary['generated_image_prompt']=[]
+            if prompt not in dictionary['prompt_collection'] and prompt not in dictionary['generated_image_prompt']:
+                dictionary['prompt_collection']=[prompt]+dictionary['prompt_collection']
+            new_size=np.array(canvas_result.image_data).shape[:2]
+            new_size=(new_size[-1],new_size[0])
+            if bg_image!=dictionary["bg_image"] :
+                dictionary["bg_image"]=bg_image
+                if bg_image!=None:
+                    imf=Image.open(bg_image).resize(new_size)
+                else:
+                    with open("lotte_animation_saver/animation_4.json") as read:
+                        url_json=json.load(read)        
+                    st_lottie(url_json) 
+                    imf=Image.open("ALL_image_formation\home_screen.jpg").resize(new_size)
+            else:
+                if len(dictionary['current_image'])!=0:
+                    imf=dictionary['current_image'][0]
+                else:
+                    with open("lotte_animation_saver/animation_4.json") as read:
+                        url_json=json.load(read)        
+                    st_lottie(url_json) 
+                    imf=Image.open("ALL_image_formation\home_screen.jpg")
+
+            negative_image =d4_to_3d(np.array(canvas_result.image_data))
+            if np.sum(negative_image)==0:
+                negative_image=Image.fromarray(np.where(negative_image == False, True, negative_image))
+            else:
+                negative_image=Image.fromarray(negative_image)
+            
+            modifiedValue=model_out_put(imf,negative_image,prompt,negative_prompt)
+            modifiedValue.save("ALL_image_formation/current_session_image.png")
+            dictionary['current_image']=[modifiedValue]+dictionary['current_image']
+            dictionary['every_prompt_with_val'].append((prompt,modifiedValue))
+            st.rerun()
+        else:
+            st.write("nothing importent")
+            modifiedValue="@working"
+            dictionary['every_prompt_with_val'].append((prompt,modifiedValue))
+            st.rerun()
+            # st.image(modifiedValue,width=300)
+    
+        
+        
+if canvas_result.json_data is not None:
+    objects = pd.json_normalize(canvas_result.json_data["objects"]) # need to convert obj to str because PyArrow
+    for col in objects.select_dtypes(include=['object']).columns:
+        objects[col] = objects[col].astype("str")
+