Update app.py

app.py

@@ -2,6 +2,7 @@
 @author: idoia lerchundi
 """
 import os
+import time
 import streamlit as st
 from huggingface_hub import InferenceClient
 import random
@@ -12,6 +13,17 @@ api_key = os.getenv("HF_TOKEN")
 # Instantiate the InferenceClient
 client = InferenceClient(api_key=api_key)
 
+# Function to simulate some process and return the elapsed time
+def process_with_timing(): start_time = time.time() 
+    # Simulate a process with sleep
+    time.sleep(2.345) 
+    # Change this value to simulate different processing times
+    end_time = time.time() 
+    elapsed_time = end_time - start_time
+    minutes, seconds = divmod(elapsed_time, 60)
+    milliseconds = (seconds - int(seconds)) * 1000
+    return minutes, int(seconds), milliseconds
+
 # Streamlit app title
 st.title("Text-generation model using Streamlit from Inference API (serverless) feature.") 
 
@@ -23,8 +35,6 @@ if "full_text" not in st.session_state:
 model_options = [ "TinyLlama/TinyLlama-1.1B-Chat-v1.0", "gpt2", "facebook/opt-1.3b", "EleutherAI/gpt-neo-2.7B","meta-llama/Llama-Llama-3-8B-Instruct", "meta-llama/Llama-Llama-3.1-1B-Instruct", "meta-llama/Llama-Llama-3.2-3B-Instruct", "meta-llama/Llama-Llama-3.2-8B-Instruct", "Qwen/Qwen2.5-1.5B-Instruct", "openai-community/gpt2", "google/gemma-1.1-7b-it", "google/gemma-1.27b-it", "google/gemma-1.2b-it", "google/gemma-1.9b-it", "google/gemma-2.2b-it", "HuggingFaceH4/starchat7b-beta", "distilbert/distilgpt2", "facebook/opt-1.3b", "distributed/optimized=gpt2-1b" ]
 selected_model = st.selectbox("Choose a model:", model_options)
 
-
-
 # Create a text input area for user prompts
 with st.form("my_form"):
     text = st.text_area("JOKER (TinyLlama is not great at joke telling.) (using model TinyLlama/TinyLlama-1.1B-Chat-v1.0):", "Tell me a clever and funny joke in exactly 4 sentences. It should make me laugh really hard. Don't repeat the topic in your joke. Be creative and concise.")
@@ -51,7 +61,9 @@ if submitted:
         top_p=random.uniform(0.7, 1.0),
         stream=True
     )
-
+    minutes, seconds, milliseconds = process_with_timing()
+    st.write(f"Elapsed Time: {int(minutes)} minutes, {seconds} seconds, and {milliseconds:.2f} milliseconds")
+    
     # Concatenate chunks to form the full response
     for chunk in stream:
         full_text += chunk.choices[0].delta.content