i

app.py

@@ -26,24 +26,22 @@ def get_completion(prompt, model="dolly-v0-70m"):
     )
 
 	# Examples from https://www.databricks.com/blog/2023/03/24/hello-dolly-democratizing-magic-chatgpt-open-models.html
-	instructions = [
-    	prompt
-		]
-
-		# set some additional pipeline args
-		pipeline_kwargs = {'torch_dtype': "auto"}
-		#if gpu_family == "v100":
-    		#pipeline_kwargs['torch_dtype'] = "float16"
-		#elif gpu_family == "a10" or gpu_family == "a100":
-    		#pipeline_kwargs['torch_dtype'] = "bfloat16"
-
-		pipeline_kwargs['max_new_tokens'] = 300
-
-		# Use the model to generate responses for each of the instructions above.
-		for instruction in instructions:
-    		response = generate_response(instruction, model=model, tokenizer=tokenizer, **pipeline_kwargs)
-    	if response:
-        	print(f"Instruction: {instruction}\n\n{response}\n\n-----------\n")
+	instructions = [prompt]
+
+	# set some additional pipeline args
+	pipeline_kwargs = {'torch_dtype': "auto"}
+	#if gpu_family == "v100":
+    #pipeline_kwargs['torch_dtype'] = "float16"
+	#elif gpu_family == "a10" or gpu_family == "a100":
+    #pipeline_kwargs['torch_dtype'] = "bfloat16"
+
+	pipeline_kwargs['max_new_tokens'] = 300
+
+	# Use the model to generate responses for each of the instructions above.
+	for instruction in instructions:
+    	response = generate_response(instruction, model=model, tokenizer=tokenizer, **pipeline_kwargs)
+    if response:
+    	print(f"Instruction: {instruction}\n\n{response}\n\n-----------\n")
 
     return response