update

DN6 · DN6 · commit 30f13ed310df · 2024-12-05T09:12:42.000+01:00
diff --git a/tests/quantization/gguf/test_gguf.py b/tests/quantization/gguf/test_gguf.py
@@ -82,6 +82,7 @@ def test_gguf_memory(self):
             self.ckpt_path, quantization_config=quantization_config, torch_dtype=self.torch_dtype
         )
         model.to("cuda")
+        assert (model.get_memory_footprint() / 1024**3) < 5
         inputs = self.get_dummy_inputs()
 
         torch.cuda.reset_peak_memory_stats()

Original file line number	Diff line number	Diff line change
`@@ -82,6 +82,7 @@ def test_gguf_memory(self):`
`82`	`82`	`self.ckpt_path, quantization_config=quantization_config, torch_dtype=self.torch_dtype`
`83`	`83`	`)`
`84`	`84`	`model.to("cuda")`
	`85`	`+ assert (model.get_memory_footprint() / 1024**3) < 5`
`85`	`86`	`inputs = self.get_dummy_inputs()`
`86`	`87`
`87`	`88`	`torch.cuda.reset_peak_memory_stats()`