How to use eval_image_classifier.py to predict my images?

[phybrain]

I'm not very familiar with this code enough，can you help me？

[pudae]

Here is sample code. link Usage: python predict_image.py --image=./sample_image.jpg --checkpoint_path=./tf-densenet121.ckpt

[mit10000]

Pudae, Thanks for sharing. One further question. In your code, the output logits is defined by

logits, _ = network_fn(images)

In deployment, if we don't want to disclose network_fn source codes, instead we use

logits=tf.get_default_graph().get_tensor_by_name('densenet169/logits/BiasAdd:0')

We can do prediction, but the results are different. Could you please tell what cause this problem?