Convertion of vectors to .bytes for standalone embedding projector

Trying to use my own vectors/labels into the standalone embedding projector instead of the example model.
I did so as a comment below suggests and with following two files:

1. vectors.bytes which is created with numpy's tofile() command:

array([[ 2.5136387 , -2.084486  ,  0.23319365],
       [ 4.008879  ,  1.2343712 , -1.1920882 ],
       [ 3.4046457 , -0.42918122, -0.84252584],
       ...,
       [ 3.2287045 , -0.49425614, -0.7298379 ],
       [ 0.9056233 , -0.4525969 ,  2.068056  ],
       [ 2.942347  ,  0.94748515, -0.8100381 ]], dtype=float32)

vectors.tofile(os.path.join(LOG_DIR, "tensors.bytes")) (array size is 5000x3)

2. meta.tsv which is a one column tab seperated file containing 5000 labels.

in standalone_projector_config.json I added

{
  "embeddings": [
    {
      "tensorName": "Doc2Vec Lipid",
      "tensorShape": [
        5000,
        3
      ],
      "tensorPath": "/PATH/tensors.bytes",
      "metadataPath": "/PATH/meta.tsv"
    },
...
}

When uploading the vectors/meta files, in .TSV file format to embedding projector (either online or standalone), everything works as intended but including them in bytes/TSV format I get the following error message:

Bazel Closure Rules</h1><h3>//tensorboard/plugins/projector/vz_projector:standalone</h3><p>No srcs found in transitive closure with path component prefix matching request path.

@harveyslash The .bytes files are binary files filled with a bunch of 32 bit floating point numbers. You can use a numpy array with dtype=numpy.float32 and tofile to generate them:

vectors = numpy.zeros(vector_shape, dtype=numpy.float32)
vectors.tofile('my_tensors.bytes')

Here's a complete example of exporting spaCy Vocab vectors for use in the standalone projector.
https://github.com/explosion/spaCy/blob/master/examples/vectors_tensorboard_standalone.py

Originally posted by @justindujardin in https://github.com/tensorflow/tensorflow/issues/7562#issuecomment-376606027