Strange result while classifying an image

[huuphucnguyen]

Hello,
I tried using the re-trained model as pointed in the 6th section of the tutorial
(https://codelabs.developers.google.com/codelabs/tensorflow-for-poets/#5),
but the image is an table downloaded from this link
https://www.target.com/p/farm-60-dining-table-honey-threshold-153/-/A-15587335,

$python -m scripts.label_image --graph=tf_files/retrained_graph.pb --image=tf_files/test_cus/15587335.jpeg
2018-09-10 09:58:20.939630: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA

Evaluation time (1-image): 0.135s

tulips (score=0.94892)
dandelion (score=0.04773)
roses (score=0.00280)
daisy (score=0.00040)
sunflowers (score=0.00015)

The result was very strange! I'm a newbie, so could you help me any explanation, please?