Removed onnxruntime models to save space

app.py

@@ -28,14 +28,14 @@ key = os.environ.get("GOOGLE_BOOKS_API_KEY")
 # Define summarization models
 summary_tokenizer_normal = AutoTokenizer.from_pretrained("lidiya/bart-base-samsum")
 summary_model_normal = AutoModelForSeq2SeqLM.from_pretrained("lidiya/bart-base-samsum")
-summary_tokenizer_onnx = AutoTokenizer.from_pretrained("optimum/t5-small")
-summary_model_onnx = ORTModelForSeq2SeqLM.from_pretrained("optimum/t5-small")
+#summary_tokenizer_onnx = AutoTokenizer.from_pretrained("optimum/t5-small")
+#summary_model_onnx = ORTModelForSeq2SeqLM.from_pretrained("optimum/t5-small")
 
 # Define classification models
-classification_tokenizer_normal = AutoTokenizer.from_pretrained(
+#classification_tokenizer_normal = AutoTokenizer.from_pretrained(
     "sileod/deberta-v3-base-tasksource-nli"
 )
-classification_model_normal = AutoModelForSequenceClassification.from_pretrained(
+#classification_model_normal = AutoModelForSequenceClassification.from_pretrained(
     "sileod/deberta-v3-base-tasksource-nli"
 )
 
@@ -502,9 +502,6 @@ async def summarize(descriptions: list, runtime="normal"):
         tokenizer = summary_tokenizer_normal
         normal_model = summary_model_normal
         model = BetterTransformer.transform(normal_model)
-    elif runtime == "onnxruntime":
-        tokenizer = summary_tokenizer_onnx
-        model = summary_model_onnx
 
     # Create the summarizer pipeline
     summarizer_pipe = pipeline("summarization", model=model, tokenizer=tokenizer)