Update app.py

app.py

@@ -37,7 +37,7 @@ def get_detailed_instruct(task_description: str, query: str) -> str:
 @spaces.GPU
 def compute_embeddings(*input_texts):
     tokenizer = AutoTokenizer.from_pretrained('intfloat/e5-mistral-7b-instruct')
-    model = AutoModel.from_pretrained('intfloat/e5-mistral-7b-instruct', torch_dtype="auto", device_map=device)
+    model = AutoModel.from_pretrained('intfloat/e5-mistral-7b-instruct', torch_dtype=torch.float16, device_map=device)
     max_length = 4096
     task = 'Given a web search query, retrieve relevant passages that answer the query'
     processed_texts = [get_detailed_instruct(task, text) for text in input_texts]