From: Jongsoo Park <jongsoo@fb.com>
Date: Mon, 7 Jan 2019 19:04:22 +0000 (-0800)
Subject: fix conv unit test for groupwise quantization and pre-packing (#15761)
X-Git-Tag: accepted/tizen/6.5/unified/20211028.231830~1996
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=c68eb5ec442c4e6cd2ad1a443ac8128b3cd99312;p=platform%2Fupstream%2Fpytorch.git

fix conv unit test for groupwise quantization and pre-packing (#15761)

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/15761

As title says.

Reviewed By: csummersea

Differential Revision: D13587727

fbshipit-source-id: f0631b8cbb89d65a1d952bc25b463de23de93bec
---

diff --git a/caffe2/quantization/server/conv_groupwise_dnnlowp_op_test.py b/caffe2/quantization/server/conv_groupwise_dnnlowp_op_test.py
index 8e73edd..e3ed280 100644
--- a/caffe2/quantization/server/conv_groupwise_dnnlowp_op_test.py
+++ b/caffe2/quantization/server/conv_groupwise_dnnlowp_op_test.py
@@ -124,7 +124,11 @@ class GroupWiseDNNLowPOpConvTest(hu.HypothesisTestCase):
 
             conv = core.CreateOperator(
                 op_type,
-                ["X_q" if do_quantize else "X", "W", "b"],
+                [
+                    "X_q" if do_quantize else "X",
+                    "W_packed" if do_prepack_weight else "W",
+                    "b",
+                ],
                 ["Y_q" if do_dequantize else "Y"],
                 stride=stride,
                 kernel=kernel,