Following Introduction to TensorFlow Hello World, I tried image recognition using learning data called ImageNet.
syntax
(tensorflow)$ python classify_image.py --image_file [Image file]
I downloaded the image of the stove with wget and executed it as follows.
wget https://qiita-image-store.s3.amazonaws.com/0/26079/ef75a5b4-8ddd-0015-9b3d-194b55a6b88e.jpeg
python classify_image.py --image_file ef75a5b4-8ddd-0015-9b3d-194b55a6b88e.jpeg
Display result
stove (score = 0.99871)
fire screen, fireguard (score = 0.00047)
space heater (score = 0.00021)
rotisserie (score = 0.00002)
table lamp (score = 0.00002)
Is the probability of being a stove 99%? smart
reference: http://qiita.com/nkjm/items/a2dada74d48b29f0e5f4
Recommended Posts