Update app.py

app.py

@@ -3,25 +3,14 @@ from transformers import AutoTokenizer, AutoModelForCausalLM
 
 model_name = "defog/sqlcoder-7b-2"
 tokenizer = AutoTokenizer.from_pretrained(model_name)
-if available_memory > 15e9:
-    # if you have atleast 15GB of GPU memory, run load the model in float16
-    model = AutoModelForCausalLM.from_pretrained(
-        model_name,
-        trust_remote_code=True,
-        torch_dtype=torch.float16,
-        device_map="auto",
-        use_cache=True,
-    )
-else:
-    # else, load in 8 bits – this is a bit slower
-    model = AutoModelForCausalLM.from_pretrained(
-        model_name,
-        trust_remote_code=True,
-        # torch_dtype=torch.float16,
-        load_in_8bit=True,
-        device_map="auto",
-        use_cache=True,
-    )
+model = AutoModelForCausalLM.from_pretrained(
+    model_name,
+    trust_remote_code=True,
+    # torch_dtype=torch.float16,
+    load_in_8bit=True,
+    device_map="auto",
+    use_cache=True,
+)
 
 prompt = """### Task
 Generate a SQL query to answer [QUESTION]{question}[/QUESTION]