Make TextEmbedder and TextClassifier tests pass on Windows

PiperOrigin-RevId: 506421383

mediapipe/tasks/cc/text/text_classifier/text_classifier_test.cc

@@ -135,28 +135,46 @@ TEST_F(TextClassifierTest, TextClassifierWithBert) {
   options->base_options.model_asset_path = GetFullPath(kTestBertModelPath);
   MP_ASSERT_OK_AND_ASSIGN(std::unique_ptr<TextClassifier> classifier,
                           TextClassifier::Create(std::move(options)));
-  MP_ASSERT_OK_AND_ASSIGN(
+
-      TextClassifierResult negative_result,
-      classifier->Classify("unflinchingly bleak and desperate"));
   TextClassifierResult negative_expected;
+  TextClassifierResult positive_expected;
+
+#ifdef _WIN32
+  negative_expected.classifications.emplace_back(Classifications{
+      /*categories=*/{
+          {/*index=*/0, /*score=*/0.956124, /*category_name=*/"negative"},
+          {/*index=*/1, /*score=*/0.043875, /*category_name=*/"positive"}},
+      /*head_index=*/0,
+      /*head_name=*/"probability"});
+  positive_expected.classifications.emplace_back(Classifications{
+      /*categories=*/{
+          {/*index=*/1, /*score=*/0.999951, /*category_name=*/"positive"},
+          {/*index=*/0, /*score=*/0.000048, /*category_name=*/"negative"}},
+      /*head_index=*/0,
+      /*head_name=*/"probability"});
+#else
   negative_expected.classifications.emplace_back(Classifications{
       /*categories=*/{
           {/*index=*/0, /*score=*/0.956317, /*category_name=*/"negative"},
           {/*index=*/1, /*score=*/0.043683, /*category_name=*/"positive"}},
       /*head_index=*/0,
       /*head_name=*/"probability"});
-  ExpectApproximatelyEqual(negative_result, negative_expected);
-
-  MP_ASSERT_OK_AND_ASSIGN(
-      TextClassifierResult positive_result,
-      classifier->Classify("it's a charming and often affecting journey"));
-  TextClassifierResult positive_expected;
   positive_expected.classifications.emplace_back(Classifications{
       /*categories=*/{
           {/*index=*/1, /*score=*/0.999945, /*category_name=*/"positive"},
           {/*index=*/0, /*score=*/0.000056, /*category_name=*/"negative"}},
       /*head_index=*/0,
       /*head_name=*/"probability"});
+#endif  // _WIN32
+
+  MP_ASSERT_OK_AND_ASSIGN(
+      TextClassifierResult negative_result,
+      classifier->Classify("unflinchingly bleak and desperate"));
+  ExpectApproximatelyEqual(negative_result, negative_expected);
+
+  MP_ASSERT_OK_AND_ASSIGN(
+      TextClassifierResult positive_result,
+      classifier->Classify("it's a charming and often affecting journey"));
   ExpectApproximatelyEqual(positive_result, positive_expected);
 
   MP_ASSERT_OK(classifier->Close());
@@ -233,12 +251,17 @@ TEST_F(TextClassifierTest, BertLongPositive) {
   TextClassifierResult expected;
   std::vector<Category> categories;
 
-// Predicted scores are slightly different on Mac OS.
+// Predicted scores are slightly different across platforms.
 #ifdef __APPLE__
   categories.push_back(
       {/*index=*/1, /*score=*/0.974181, /*category_name=*/"positive"});
   categories.push_back(
       {/*index=*/0, /*score=*/0.025819, /*category_name=*/"negative"});
+#elif defined _WIN32
+  categories.push_back(
+      {/*index=*/1, /*score=*/0.976686, /*category_name=*/"positive"});
+  categories.push_back(
+      {/*index=*/0, /*score=*/0.023313, /*category_name=*/"negative"});
 #else
   categories.push_back(
       {/*index=*/1, /*score=*/0.985889, /*category_name=*/"positive"});

mediapipe/tasks/cc/text/text_embedder/text_embedder_test.cc

@@ -75,7 +75,11 @@ TEST_F(EmbedderTest, SucceedsWithMobileBert) {
       text_embedder->Embed("it's a charming and often affecting journey"));
   ASSERT_EQ(result0.embeddings.size(), 1);
   ASSERT_EQ(result0.embeddings[0].float_embedding.size(), 512);
+#ifdef _WIN32
+  ASSERT_NEAR(result0.embeddings[0].float_embedding[0], 21.2148f, kEpsilon);
+#else
   ASSERT_NEAR(result0.embeddings[0].float_embedding[0], 19.9016f, kEpsilon);
+#endif  // _WIN32
 
   MP_ASSERT_OK_AND_ASSIGN(
       auto result1, text_embedder->Embed("what a great and fantastic trip"));
@@ -87,7 +91,11 @@ TEST_F(EmbedderTest, SucceedsWithMobileBert) {
   MP_ASSERT_OK_AND_ASSIGN(
       double similarity, TextEmbedder::CosineSimilarity(result0.embeddings[0],
                                                         result1.embeddings[0]));
+#ifdef _WIN32
+  EXPECT_NEAR(similarity, 0.971417, kSimilarityTolerancy);
+#else
   EXPECT_NEAR(similarity, 0.969514, kSimilarityTolerancy);
+#endif  // _WIN32
 
   MP_ASSERT_OK(text_embedder->Close());
 }
@@ -160,8 +168,12 @@ TEST_F(EmbedderTest, SucceedsWithMobileBertAndDifferentThemes) {
   MP_ASSERT_OK_AND_ASSIGN(
       double similarity, TextEmbedder::CosineSimilarity(result0.embeddings[0],
                                                         result1.embeddings[0]));
-  // TODO: The similarity should likely be lower
+  // TODO: These similarity should likely be lower
+#ifdef _WIN32
+  EXPECT_NEAR(similarity, 0.98152, kSimilarityTolerancy);
+#else
   EXPECT_NEAR(similarity, 0.98088, kSimilarityTolerancy);
+#endif  // _WIN32
 
   MP_ASSERT_OK(text_embedder->Close());
 }