Update app.py
Browse files
app.py
CHANGED
|
@@ -79,7 +79,7 @@ quantization_config = BitsAndBytesConfig(load_in_8bit=True)
|
|
| 79 |
model = AutoModelForCausalLM.from_pretrained(
|
| 80 |
"m-a-p/YuE-s1-7B-anneal-en-cot",
|
| 81 |
torch_dtype="auto",
|
| 82 |
-
attn_implementation="flash_attention_2",
|
| 83 |
quantization_config=quantization_config,
|
| 84 |
low_cpu_mem_usage=True,
|
| 85 |
device_map="auto"
|
|
|
|
| 79 |
model = AutoModelForCausalLM.from_pretrained(
|
| 80 |
"m-a-p/YuE-s1-7B-anneal-en-cot",
|
| 81 |
torch_dtype="auto",
|
| 82 |
+
# attn_implementation="flash_attention_2",
|
| 83 |
quantization_config=quantization_config,
|
| 84 |
low_cpu_mem_usage=True,
|
| 85 |
device_map="auto"
|