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 ... ### Environment information ``` OS: $ pip freeze | grep tensor # your output here $ python -V # your output here ``` ### For bugs: reproduction...
### Description How can I add next_frame_glow to the registry? It is not on the list (nor are the corresponding hyperparameters) when I run t2t_trainer --registry_help
What script to run in order to compute the approx_bleu on a specific checkpoint? (Same thing that is done during training every X steps)
Hello , I use T2T for translation task, it's version is 1.5.5. The setting I use as follows: PROBLEM=translate_enzh_wmt32k MODEL=transformer HPARAMS=transformer_base_single_gpu I used t2t-trainer.py to train a model. When it...
Fixed Common Voice data generator by adding a flag to the `datagen.py` file for the language code (`--language="en"`, if not specified it defaults to english) and dynamically downloading the correct...
### Description When working with t2t 1.15.7 on tensorflow 2.2 and performing training on 1 GPU the model weights are ~211M, but when we increase the # of GPUs the...
### Description Hi, I am getting the file not found error whenever I run t2t_problem.generate_data(DATA_DIR, TMP_DIR). Download of the data set was successful to TMP_DIR. ... ### Environment information Python...
I am getting an OoM error while training with 8 GPUs but not with 1 GPU. I use the following command to train. t2t-trainer \ --data_dir=$DATA_DIR \ --problem=$PROBLEM \ --model=$MODEL...
Hello I got the errors about jaxlib when I was trying to install tensor2tensor on windows. The error message was blow ERROR: Could not find a version that satisfies the...
### Description I am trying to run transformer using 1 worker and 1 ps in async mode. The program hangs after printing `INFO:tensorflow:Graph was finalized`. ### Environment information ``` OS:...
Metadata
Owner
Metadata
Library of deep learning models and datasets designed to make deep learning more accessible and accelerate ML research.