fold quantize in convert

mcr229 · web-flow · commit 5395ae696403 · 2024-08-27T18:32:22.000-07:00
Differential Revision: D61814397 Pull Request resolved: #4889
diff --git a/examples/models/phi-3-mini/export_phi-3-mini.py b/examples/models/phi-3-mini/export_phi-3-mini.py
@@ -69,7 +69,7 @@ def export(args) -> None:
         )
         model = prepare_pt2e(model, xnnpack_quantizer)  # pyre-fixme[6]
         model(*example_inputs)
-        model = convert_pt2e(model, fold_quantize=False)
+        model = convert_pt2e(model)
         DuplicateDynamicQuantChainPass()(model)
         # TODO(lunwenh): update it to use export once
         # https://github.com/pytorch/pytorch/issues/128394 is resolved.

Original file line number	Diff line number	Diff line change
`@@ -69,7 +69,7 @@ def export(args) -> None:`
`69`	`69`	`)`
`70`	`70`	`model = prepare_pt2e(model, xnnpack_quantizer) # pyre-fixme[6]`
`71`	`71`	`model(*example_inputs)`
`72`		`- model = convert_pt2e(model, fold_quantize=False)`
	`72`	`+ model = convert_pt2e(model)`
`73`	`73`	`DuplicateDynamicQuantChainPass()(model)`
`74`	`74`	`# TODO(lunwenh): update it to use export once`
`75`	`75`	`# https://github.com/pytorch/pytorch/issues/128394 is resolved.`