We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent 1cf886b commit 0083f9cCopy full SHA for 0083f9c
tests/test_gpu_examples.py
@@ -2199,8 +2199,8 @@ class TestLoftQ:
2199
# The error factor indicates by how much the quantization error should be decreased when using LoftQ compared to
2200
# quantization without LoftQ. Thus 1.03 means that the error should be decreased by 3% at least. This is a very
2201
# conservative value to prevent flakiness, in practice most gains are > 1.5
2202
- error_factor = 1.03
2203
device = infer_device()
+ error_factor = 1.005 if device in ("xpu", "cpu") else 1.03
2204
2205
def get_input(self, model_id, device):
2206
tokenizer = AutoTokenizer.from_pretrained(model_id)
0 commit comments