Convert between NHWC and NCHW in TensorFlow

by

All you need to do is a permutation of the dimensions from NHWC to NCHW (or the contrary).

The meaning of each letter might help understand:

  • N: number of images in the batch
  • H: height of the image
  • W: width of the image
  • C: number of channels of the image (ex: 3 for RGB, 1 for grayscale…)

From NHWC to NCHW

The image shape is (N, H, W, C) and we want the output to have shape (N, C, H, W). Therefore we need to apply tf.transpose with a well chosen permutation perm.

The returned tensor’s dimension i will correspond to the input dimension perm[i]

perm[0] = 0  # output dimension 0 will be 'N', which was dimension 0 in the input
perm[1] = 3  # output dimension 1 will be 'C', which was dimension 3 in the input
perm[2] = 1  # output dimension 2 will be 'H', which was dimension 1 in the input
perm[3] = 2  # output dimension 3 will be 'W', which was dimension 2 in the input

In practice:

images_nhwc = tf.placeholder(tf.float32, [None, 200, 300, 3])  # input batch
out = tf.transpose(images_nhwc, [0, 3, 1, 2])
print(out.get_shape())  # the shape of out is [None, 3, 200, 300]

From NCHW to NHWC

The image shape is (N, C, H, W) and we want the output to have shape (N, H, W, C). Therefore we need to apply tf.transpose with a well chosen permutation perm.

The returned tensor’s dimension i will correspond to the input dimension perm[i]

perm[0] = 0  # output dimension 0 will be 'N', which was dimension 0 in the input
perm[1] = 2  # output dimension 1 will be 'H', which was dimension 2 in the input
perm[2] = 3  # output dimension 2 will be 'W', which was dimension 3 in the input
perm[3] = 1  # output dimension 3 will be 'C', which was dimension 1 in the input

In practice:

images_nchw = tf.placeholder(tf.float32, [None, 3, 200, 300])  # input batch
out = tf.transpose(images_nchw, [0, 2, 3, 1])
print(out.get_shape())  # the shape of out is [None, 200, 300, 3]

More Related Contents:

  1. How to compile Tensorflow with SSE4.2 and AVX instructions?
  2. On Windows, running “import tensorflow” generates No module named “_pywrap_tensorflow” error
  3. Meaning of buffer_size in Dataset.map , Dataset.prefetch and Dataset.shuffle
  4. How does TensorFlow name tensors?
  5. What is the meaning of the word logits in TensorFlow? [duplicate]
  6. How to set specific gpu in tensorflow?
  7. Can I use TensorBoard with Google Colab?
  8. How do I split Tensorflow datasets?
  9. After building TensorFlow from source, seeing libcudart.so and libcudnn errors
  10. Multivariate LSTM with missing values
  11. TensorFlow: numpy.repeat() alternative
  12. Specify connections in NN (in keras)
  13. What are the differences between all these cross-entropy losses in Keras and TensorFlow?
  14. Tensorflow: ImportError: libcusolver.so.8.0: cannot open shared object file: No such file or directory
  15. Tensorflow crashes with CUBLAS_STATUS_ALLOC_FAILED
  16. Keras Loss Function with Additional Dynamic Parameter
  17. TensorFlow: How to measure how much GPU memory each tensor takes?
  18. How to convert .pb to TFLite format?
  19. How to select rows from a 3-D Tensor in TensorFlow?
  20. Working with multiple graphs in TensorFlow
  21. ValueError: Shapes (None, 1) and (None, 2) are incompatible
  22. Update only part of the word embedding matrix in Tensorflow
  23. Could not install packages due to an EnvironmentError: [WinError 5] Access is denied:
  24. CUDA_ERROR_OUT_OF_MEMORY in tensorflow
  25. Implementing contrastive loss and triplet loss in Tensorflow
  26. Confusion about keras Model: __call__ vs. call vs. predict methods
  27. What does batch, repeat, and shuffle do with TensorFlow Dataset?
  28. How to replace (or insert) intermediate layer in Keras model?
  29. Choosing number of Steps per Epoch
  30. Parallelism isn’t reducing the time in dataset map

Leave a Comment