bigscience-workshop/t-zero
Reproduce results and replicate training fo T0 (Multitask Prompted Training Enables Zero-Shot Task Generalization)
PythonApache-2.0
Issues
- 4
ETA on pytorch training code?
#13 opened by NtaylorOX - 1
fine tuning on other LM models
#43 opened by ShiJiawenwen - 1
- 2
mesh tensorflow checkpoints?
#25 opened by hamishivi - 4
Truncation method
#39 opened by gportill - 5
Prediction for multi-token multiple choice?
#38 opened by AADeLucia - 3
Evaluation on ANLI fails with `ValueError`
#37 opened by fepegar - 1
T0 (p=1) replicability
#35 opened by tuhinjubcse - 1
- 2
- 1
Training duration
#26 opened by kevinpl07 - 1
big-bench evaluation
#30 opened by thesofakillers - 2
Missing dataset config names in P3
#31 opened by ari9dam - 4
how to use this t-zero to train my own dataset? did not find the api . if add the dataset to the cache_data_dir.I do not want reproduce. I want to use this tool train my own datasets
#20 opened by flyingwaters - 1
how to reproduce the result in offline environment , I dowload the sentence.model and checkpoint but, the sentencepiece.model can not be recognised by the t5~!!
#21 opened by flyingwaters - 0
how
#19 opened by flyingwaters - 2
- 1
- 2
- 2
- 1
reproducibility versus replicability
#7 opened by valdanchev