search

omoindrot / tensorflow-triplet-loss

Implementation of triplet loss in TensorFlow
https://omoindrot.github.io/triplet-loss
MIT License
1.12k stars 283 forks source link

How to get the embedding of a single image? #43

Closed agi-hub closed 5 years ago

agi-hub commented 5 years ago

I am trying to inference using the trained network, however I am not sure where to put the image in, and what is the network node of the embedding. Since the estimator with dataset is not friendly for a raw image, could you provide an interface more easy for inference?

agi-hub commented 5 years ago

I tried the method below by using the estimator you gave in the visualize_embeddings.py.

` mnist = tf.keras.datasets.mnist

(x_train, y_train),(x_test, y_test) = mnist.load_data() x_train, x_test = x_train / 255.0, x_test / 255.0

predict_input_fn = tf.estimator.inputs.numpy_input_fn( x={"x": x_test[1][np.newaxis, :, :]}, num_epochs=1, shuffle=False)

y = estimator.predict(input_fn=predict_input_fn)

` I can only get the y as an object shown in pdb: <generator object Estimator.predict at 0x7f2f5d0a46d0>

Then what to do next to obtain the data?

omoindrot commented 5 years ago

estimator.predict(...) returns a generator that you can iterate on:

generator = estimator.predict(input_fn=predict_input_fn)
for pred in generator:
    ...