Messing with configuration.

app.py

@@ -32,11 +32,20 @@ print(f"CUDA device: {torch.cuda.get_device_name(torch.cuda.current_device())}")
 # "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"
-file_name = "onnx/model.onnx"
-tokenizer_name = "MoritzLaurer/deberta-v3-base-zeroshot-v2.0"
-model = ORTModelForSequenceClassification.from_pretrained(model_name, file_name=file_name, provider="CUDAExecutionProvider")
-tokenizer = AutoTokenizer.from_pretrained(tokenizer_name, model_max_length=512)
+# model_name = "MoritzLaurer/deberta-v3-base-zeroshot-v2.0"
+# file_name = "onnx/model.onnx"
+# tokenizer_name = "MoritzLaurer/deberta-v3-base-zeroshot-v2.0"
+# model = ORTModelForSequenceClassification.from_pretrained(model_name, file_name=file_name, provider="CUDAExecutionProvider")
+# tokenizer = AutoTokenizer.from_pretrained(tokenizer_name, model_max_length=512)
+
+model = ORTModelForSequenceClassification.from_pretrained(
+    "philschmid/tiny-bert-sst2-distilled",
+    export=True,
+    provider="CUDAExecutionProvider",
+)
+
+tokenizer = AutoTokenizer.from_pretrained("philschmid/tiny-bert-sst2-distilled")
+
 classifier = pipeline(task="zero-shot-classification", model=model, tokenizer=tokenizer, device="cuda:0")
 
 def classify(data_string, request: gradio.Request):