Github user imatiach-msft commented on a diff in the pull request: https://github.com/apache/spark/pull/20583#discussion_r167471442 --- Diff: mllib/src/test/scala/org/apache/spark/ml/image/ImageSchemaSuite.scala --- @@ -65,11 +65,71 @@ class ImageSchemaSuite extends SparkFunSuite with MLlibTestSparkContext { assert(count50 > 0 && count50 < countTotal) } + test("readImages test: recursive = false") { + val df = readImages(imagePath, null, false, 3, true, 1.0, 0) + assert(df.count() === 0) + } + + test("readImages test: read jpg image") { + val df = readImages(imagePath + "/kittens/DP153539.jpg", null, false, 3, true, 1.0, 0) + assert(df.count() === 1) + } + + test("readImages test: read png image") { + val df = readImages(imagePath + "/multi-channel/BGRA.png", null, false, 3, true, 1.0, 0) + assert(df.count() === 1) + } + + test("readImages test: read non image") { + val df = readImages(imagePath + "/kittens/not-image.txt", null, false, 3, true, 1.0, 0) + assert(df.count() === 0) + } + + test("readImages test: read non image and dropImageFailures is false") { + val df = readImages(imagePath + "/kittens/not-image.txt", null, false, 3, false, 1.0, 0) + assert(df.count() === 1) + } + + test("readImages test: sampleRatio > 1") { + val e = intercept[IllegalArgumentException] { + readImages(imagePath, null, true, 3, true, 1.1, 0) + } + assert(e.getMessage.equals("requirement failed: sampleRatio should be between 0 and 1")) + } + + test("readImages test: sampleRatio < 0") { + val e = intercept[IllegalArgumentException] { + readImages(imagePath, null, true, 3, true, -0.1, 0) + } + assert(e.getMessage.equals("requirement failed: sampleRatio should be between 0 and 1")) + } + + test("readImages test: sampleRatio = 0") { + val df = readImages(imagePath, null, true, 3, true, 0.0, 0) + assert(df.count() === 0) + } + + test("readImages test: with sparkSession") { + val df = readImages(imagePath, sparkSession = spark, true, 3, true, 1.0, 0) --- End diff -- it looks like this test is failing in the latest build? I'm not sure why, the error message says the count is 8. This test, which is similar, seems to have a count of 7 ("readImages count test"): df = readImages(imagePath, null, true, -1, true, 1.0, 0)
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org