tensor2tensor
tensor2tensor copied to clipboard
tensorflow
Library of deep learning models and datasets designed to make deep learning more accessible and accelerate ML research.
### Description - Trained translation problem transformer_tiny model. - Exported it and loaded for serving using t2t-exporter - Started tensorflow/serving docker to serve the exported model - converted input string...
Are there any helpful posts or requirements into how to use tensor2tensor with a custom vocabulary? It's for a translation problem. For example, do we need to include and as...
I'm new to neural architecture search. Thank you.
### Description The checkpoint is always not found in T2T-decoder. After deleting checkpoint, can run T2T-decoder,Which model is used at this time?and why? ### Environment information OS: $ pip freeze...
### Description I implemented a custom translation decoding algorithm using transformer. But I encounter a problem when I want to export my model. Is it possible to pass some python...
Quote and single quote are not handled correctly in vocab file where words are not wrapped in quotes
Especially following branch will remove the quote so that it becomes empty string (as single quote character starts and ends with quote): https://github.com/tensorflow/tensor2tensor/blob/5f9dd2db6d7797162e53adf152310ed13e9fc711/tensor2tensor/data_generators/text_encoder.py#L929 easy fix is the check also that...
kfac > 0.2 requires tensorflow-probability==0.8.0. However, T2T requires tf-probability==0.7.0. Restricting kfac to ==0.1.5 so that it can work.
### Description when i use tst-decoder， i want to get each timestep's probablity, such as 'a' [0.1, 0.1, 0.2.....]。i tried to add log parameter in decode_hparams， it didn't work。 i...
### Description I've trained a model to rewrite a sentence to different way. every time i predict a sentence the model give me the same answer. is it possible to...
In https://github.com/tensorflow/tensor2tensor/blob/master/tensor2tensor/layers/common_attention.py  `x_t_r` shape is `[sequence_length, batch_size, hidden_size]` `z` shape is `[sequence_length, sequence_length, hidden_size]` then `tf.matmul` `x_t_r` and `z` to get a tensor with shape `[batch_size, sequence_length, sequence_length]`, How...
Metadata
Owner
Metadata
Library of deep learning models and datasets designed to make deep learning more accessible and accelerate ML research.