search

tensorlab / tensorfx

TensorFlow framework for training and serving machine learning models
Apache License 2.0
196 stars 41 forks source link

Implement a schema field that can represent multi-valent value #20

Open nikhilk opened 7 years ago

nikhilk commented 7 years ago

In tf.Example, this is intrinsically supported by a feature with N entries in a list field. Think about what is the best way to support this in csv.

Couple of options come to mind:

  1. A ':' delimited set of numbers in a column. (pros: easy to produce; cons: requires more expensive string splitting)
  2. A sequence of bytes encoded as base64 into a column. (pros: more painful to produce; still requires base64 decoding but cheaper)

Scenarios

  1. Text run through a word2vec transform during data preparation.
  2. Image run through a image2vec (aka use inception model to extract features from image)