How to display custom images in TensorBoard using Keras?

by

So, the following solution works well for me:

import tensorflow as tf

def make_image(tensor):
    """
    Convert an numpy representation image to Image protobuf.
    Copied from https://github.com/lanpa/tensorboard-pytorch/
    """
    from PIL import Image
    height, width, channel = tensor.shape
    image = Image.fromarray(tensor)
    import io
    output = io.BytesIO()
    image.save(output, format="PNG")
    image_string = output.getvalue()
    output.close()
    return tf.Summary.Image(height=height,
                         width=width,
                         colorspace=channel,
                         encoded_image_string=image_string)

class TensorBoardImage(keras.callbacks.Callback):
    def __init__(self, tag):
        super().__init__() 
        self.tag = tag

    def on_epoch_end(self, epoch, logs={}):
        # Load image
        img = data.astronaut()
        # Do something to the image
        img = (255 * skimage.util.random_noise(img)).astype('uint8')

        image = make_image(img)
        summary = tf.Summary(value=[tf.Summary.Value(tag=self.tag, image=image)])
        writer = tf.summary.FileWriter('./logs')
        writer.add_summary(summary, epoch)
        writer.close()

        return

tbi_callback = TensorBoardImage('Image Example')

Just pass the callback to fit or fit_generator.

Note that you can also run some operations using the model inside the callback. For example, you may run the model on some images to check its performance.

screen

More Related Contents:

  1. Keras : How should I prepare input data for RNN?
  2. Obtaining output of an Intermediate layer in TensorFlow/Keras
  3. Keras split train test set when using ImageDataGenerator
  4. Choosing number of Steps per Epoch
  5. TensorFlow and person recognition in video stream
  6. Keras, How to get the output of each layer?
  7. How to Properly Combine TensorFlow’s Dataset API and Keras?
  8. Can I use TensorBoard with Google Colab?
  9. Specify connections in NN (in keras)
  10. What are the differences between all these cross-entropy losses in Keras and TensorFlow?
  11. Keras custom loss function: Accessing current input pattern
  12. Does model.compile() initialize all the weights and biases in Keras (tensorflow backend)?
  13. Dimension of shape in conv1D
  14. Keras Loss Function with Additional Dynamic Parameter
  15. Using Gensim Fasttext model with LSTM nn in keras
  16. Simple way to visualize a TensorFlow graph in Jupyter?
  17. ValueError: Shapes (None, 1) and (None, 2) are incompatible
  18. Multiple outputs in Keras
  19. ValueError: Input 0 of layer sequential is incompatible with the layer: : expected min_ndim=4, found ndim=3. Full shape received: [8, 28, 28]
  20. How to implement Grad-CAM on a trained network
  21. Implementing contrastive loss and triplet loss in Tensorflow
  22. Confusion about keras Model: __call__ vs. call vs. predict methods
  23. How to fix “ResourceExhaustedError: OOM when allocating tensor”
  24. What is the difference between keras and tf.keras?
  25. How do I plot a Keras/Tensorflow subclassing API model?
  26. Strange behaviour of the loss function in keras model, with pretrained convolutional base
  27. TensorFlow 2.0 Keras: How to write image summaries for TensorBoard
  28. Multi-input Multi-output Model with Keras Functional API
  29. Checkpointing keras model: TypeError: can’t pickle _thread.lock objects
  30. How to replace (or insert) intermediate layer in Keras model?

Leave a Comment