Update app.py

app.py

@@ -1,91 +1,150 @@
 import numpy as np
 import streamlit as st
-import torch
-from transformers import AutoModelForCausalLM, AutoTokenizer
+from openai import OpenAI
 import os
+import sys
+from dotenv import load_dotenv, dotenv_values
+load_dotenv()
+
+# initialize the client
+client = OpenAI(
+  base_url="https://api-inference.huggingface.co/v1",
+  api_key=os.environ.get('HUGGINGFACEHUB_API_TOKEN')  # Replace with your token
+)
+
+# Create supported models
+model_links = {
+    "Mistral-Nemo-Base-2407": "joermd/speedy-llama2",
+}
+
+#Random dog images for error message
+random_dog = ["0f476473-2d8b-415e-b944-483768418a95.jpg",
+              "1bd75c81-f1d7-4e55-9310-a27595fa8762.jpg",
+              "526590d2-8817-4ff0-8c62-fdcba5306d02.jpg",
+              "1326984c-39b0-492c-a773-f120d747a7e2.jpg",
+              "42a98d03-5ed7-4b3b-af89-7c4876cb14c3.jpg",
+              "8b3317ed-2083-42ac-a575-7ae45f9fdc0d.jpg",
+              "ee17f54a-83ac-44a3-8a35-e89ff7153fb4.jpg",
+              "027eef85-ccc1-4a66-8967-5d74f34c8bb4.jpg",
+              "08f5398d-7f89-47da-a5cd-1ed74967dc1f.jpg",
+              "0fd781ff-ec46-4bdc-a4e8-24f18bf07def.jpg",
+              "0fb4aeee-f949-4c7b-a6d8-05bf0736bdd1.jpg",
+              "6edac66e-c0de-4e69-a9d6-b2e6f6f9001b.jpg",
+              "bfb9e165-c643-4993-9b3a-7e73571672a6.jpg"]
+
 
-# Random dog images for error messages
-random_dog = [
-    "0f476473-2d8b-415e-b944-483768418a95.jpg",
-    "1bd75c81-f1d7-4e55-9310-a27595fa8762.jpg",
-    # Add more images as needed
-]
 
-# Function to reset conversation
 def reset_conversation():
-    '''Resets conversation'''
+    '''
+    Resets Conversation
+    '''
     st.session_state.conversation = []
     st.session_state.messages = []
     return None
+    
+
+
+
+# Define the available models
+models =[key for key in model_links.keys()]
+
+# Create the sidebar with the dropdown for model selection
+selected_model = st.sidebar.selectbox("Select Model", models)
+
+# Create a temperature slider
+temp_values = st.sidebar.slider('Select a temperature value', 0.0, 1.0, (0.5))
 
-# Sidebar controls
-temp_values = st.sidebar.slider('Select a temperature value', 0.0, 1.0, 0.5)
-max_token_value = st.sidebar.slider('Select a max_token value', 1000, 9000, 5000)
-st.sidebar.button('Reset Chat', on_click=reset_conversation)
+# Create a max_token slider
+max_token_value = st.sidebar.slider('Select a max_token value', 1000, 9000, (5000))
+
+#Add reset button to clear conversation
+st.sidebar.button('Reset Chat', on_click=reset_conversation) #Reset button
+
+
+# Create model description
+st.sidebar.write(f"You're now chatting with **{selected_model}**")
+st.sidebar.markdown("*Generated content may be inaccurate or false.*")
+# st.sidebar.markdown("\n[TypeGPT](https://typegpt.net).")
+
+
+
+
+if "prev_option" not in st.session_state:
+    st.session_state.prev_option = selected_model
+
+if st.session_state.prev_option != selected_model:
+    st.session_state.messages = []
+    # st.write(f"Changed to {selected_model}")
+    st.session_state.prev_option = selected_model
+    reset_conversation()
+
+
+
+#Pull in the model we want to use
+repo_id = model_links[selected_model]
+
+
+st.subheader(f'{selected_model}')
+# # st.title(f'ChatBot Using {selected_model}')
+
+# Set a default model
+if selected_model not in st.session_state:
+    st.session_state[selected_model] = model_links[selected_model] 
 
 # Initialize chat history
 if "messages" not in st.session_state:
     st.session_state.messages = []
 
+
 # Display chat messages from history on app rerun
 for message in st.session_state.messages:
     with st.chat_message(message["role"]):
         st.markdown(message["content"])
 
-# Set cache directory path to /data
-cache_dir = "/data"  # المسار المحدد للتخزين في مساحة Hugging Face
 
-# Load model and tokenizer on-demand to save memory
-if prompt := st.chat_input(f"مرحبا انا سبيدي , كيف استطيع مساعدتك ؟"):
+
+# Accept user input
+if prompt := st.chat_input(f"Hi I'm {selected_model}, ask me a question"):
+    # Display user message in chat message container
     with st.chat_message("user"):
         st.markdown(prompt)
+    # Add user message to chat history
     st.session_state.messages.append({"role": "user", "content": prompt})
     
-    # Load model only when user submits a prompt
-    try:
-        # Load the tokenizer and model with caching in the specified directory
-        tokenizer = AutoTokenizer.from_pretrained("joermd/speedy-llama2", cache_dir=cache_dir)
-        model = AutoModelForCausalLM.from_pretrained(
-            "joermd/speedy-llama2", 
-            cache_dir=cache_dir,
-            torch_dtype=torch.bfloat16,
-            device_map="auto"
-        )
-        
-        # Prepare the system message and conversation
-        system_message = {
-            "role": "system",
-            "content": "You are a friendly chatbot who answers questions in Arabic."
-        }
-        messages = [system_message, {"role": "user", "content": prompt}]
-        
-        # Create conversation prompt using chat template
-        conversation = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
-        
-        # Generate response
-        inputs = tokenizer(conversation, return_tensors="pt")
-        outputs = model.generate(
-            inputs.input_ids,
-            max_new_tokens=max_token_value,
-            temperature=temp_values,
-            do_sample=True,
-            top_k=50,
-            top_p=0.95
-        )
-        assistant_response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-        
-        # Clear memory (for CUDA) and delete the model to free up RAM
-        if torch.cuda.is_available():
-            torch.cuda.empty_cache()
-        del model
-    
-    except Exception as e:
-        assistant_response = "😵‍💫 عذراً، حدث خطأ في الاتصال! حاول مرة أخرى لاحقاً. إليك صورة كلب 🐶:"
-        st.image(f'https://random.dog/{random_dog[np.random.randint(len(random_dog))]}')
-        st.write("رسالة الخطأ:")
-        st.write(e)
-    
-    # Display assistant response
+
+    # Display assistant response in chat message container
     with st.chat_message("assistant"):
-        st.markdown(assistant_response)
-    st.session_state.messages.append({"role": "assistant", "content": assistant_response})
+        try:
+            stream = client.chat.completions.create(
+                model=model_links[selected_model],
+                messages=[
+                    {"role": m["role"], "content": m["content"]}
+                    for m in st.session_state.messages
+                ],
+                temperature=temp_values,#0.5,
+                stream=True,
+                max_tokens=max_token_value,
+            )
+    
+            response = st.write_stream(stream)
+
+        except Exception as e:
+            # st.empty()
+            response = "😵‍💫 Looks like someone unplugged something!\
+                    \n Either the model space is being updated or something is down.\
+                    \n\
+                    \n Try again later. \
+                    \n\
+                    \n Here's a random pic of a 🐶:"
+            st.write(response)
+            random_dog_pick = 'https://random.dog/'+ random_dog[np.random.randint(len(random_dog))]
+            st.image(random_dog_pick)
+            st.write("This was the error message:")
+            st.write(e)
+
+
+
+
+            
+
+    st.session_state.messages.append({"role": "assistant", "content": response})