switch models

app.py

@@ -3,6 +3,7 @@ import gradio as gr
 from pages.arena import create_arena
 from pages.summarization_playground import create_summarization_interface
 from pages.leaderboard import create_leaderboard
+from pages.batch_evaluation import create_batch_evaluation_interface
 
 def welcome_message():
     return """
@@ -27,7 +28,7 @@ with gr.Blocks() as demo:
             with gr.TabItem("Leaderboard"):
                 create_leaderboard()
             with gr.TabItem("Batch_Evaluation"):
-                create_arena()
+                create_batch_evaluation_interface()
             with gr.TabItem("Demo_of_Streaming"):
                 create_arena()
 

pages/summarization_playground.py

@@ -5,8 +5,12 @@ import random
 from utils.model import Model
 from utils.data import dataset
 
+import gc
+import torch
+
 load_dotenv()
 
+__model_on_gpu__ = ''
 model = {model_name: Model(model_name) for model_name in Model.__model_list__}
 
 random_label = '🔀 Random dialogue from dataset'
@@ -28,7 +32,17 @@ Back in Boston, Kidd is going to rely on Lively even more. He'll play close to 3
 }
 
 def generate_answer(sources, model, model_name, prompt):
-    content = prompt + '\n' + sources
+    content = prompt + '\n' + sources + '\n\n'
+    global __model_on_gpu__
+
+    if __model_on_gpu__ != model_name:
+        model[__model_on_gpu__].cpu()
+        gc.collect()
+        torch.cuda.empty_cache()
+
+        model[model_name].gpu()
+        __model_on_gpu__ = model_name
+
     answer = model[model_name].gen(content)
 
     return answer