More cleanup.

app.py

@@ -9,7 +9,7 @@ from transformers import pipeline
 from fastapi import FastAPI
 from fastapi.middleware.cors import CORSMiddleware
 
-# CORS Config
+# CORS Config - This isn't actually working; instead, I am taking a gross approach to origin whitelisting within the service.
 app = FastAPI()
 
 app.add_middleware(
@@ -24,11 +24,6 @@ print(f"Is CUDA available: {torch.cuda.is_available()}")
 print(f"CUDA device: {torch.cuda.get_device_name(torch.cuda.current_device())}")
 
 # "xenova/mobilebert-uncased-mnli" "typeform/mobilebert-uncased-mnli" Fast but small--same as bundled in Statosphere
-# "xenova/deberta-v3-base-tasksource-nli" Not impressed
-# "Xenova/bart-large-mnli" A bit slow
-# "Xenova/distilbert-base-uncased-mnli" "typeform/distilbert-base-uncased-mnli" Bad answers
-# "Xenova/deBERTa-v3-base-mnli" "MoritzLaurer/DeBERTa-v3-base-mnli" Still a bit slow and not great answers
-# "xenova/nli-deberta-v3-small" "cross-encoder/nli-deberta-v3-small" Was using this for a good while and it was...okay
 
 model_name = "MoritzLaurer/deberta-v3-base-zeroshot-v2.0"
 tokenizer_name = "MoritzLaurer/deberta-v3-base-zeroshot-v2.0"
@@ -42,13 +37,10 @@ def classify(data_string, request: gradio.Request):
             return "{}"
     data = json.loads(data_string)
 
-    # Prevent batch suggestion warning in log.
+    # Try to prevent batch suggestion warning in log.
     classifier_cpu.call_count = 0
     classifier_gpu.call_count = 0
 
-    # if 'task' in data and data['task'] == 'few_shot_classification':
-    #     return few_shot_classification(data)
-    # else:
     start_time = time.time()
     result = {}
     try:
@@ -75,10 +67,11 @@ def create_sequences(data):
 gradio_interface = gradio.Interface(
     fn = classify,
     inputs = gradio.Textbox(label="JSON Input"),
-    outputs = gradio.Textbox()
+    outputs = gradio.Textbox(),
+    title = "Statosphere Backend",
+    description = "This Space is a classification service for a set of chub.ai stages and not really intended for use through this UI."
 )
 
 app.mount("/gradio", gradio_interface)
 
-# app = gradio.mount_gradio_app(app, gradio_interface, path="/gradio")
 gradio_interface.launch()