-
```shell
Unsupported ops: Counter({'broadcast_matmul': 89, 'scalar_div': 4, 'gather': 4, 'elementwise_minimum': 3, 'fill_': 2, 'scalar_logical_less': 2, 'where': 2, 'scalar_logical_greater': 1})
```
-
first question - I saw this mentioned in the docs. How to set it up?
second question …
-
Currently mT5 is not fully supported, yielding multiple errors of type:
[ERROR] cannot find key 'encoder.block.0.layer.1.DenseReluDense.wi_0.weight'
when passing a mT5 model to the t5_ckpt_convert.p…
-
Hey guys,
Thanks for open-sourcing mT5!
Is it maybe possible to provide a link to the sentence piece model file that is used for the pre-trained mT5 checkpoints?
I could not find the file anywhe…
-
@jiazhihao What's the current state of our support for Hugging Face models?
-
Right now, we cannot set/use a custom path for the Translation pipeline. The path set in the parameter is only used as a default fallback in the event that the language pair doesn't have a Helsinki OP…
-
@versae I've been seeing in the mt5 tokenizer the extra 100 ids needed for the sentinel tokens come already included in the tokenizer.
`'gs://t5-data/vocabs/mc4.250000.100extra/sentencepiece.model'`…
-
It seems that the current version only supports T5 v1.0 (shared word embedding version)，any plan to support T5 v1.1?
-
The package is really good, but I want to use it for other language, and I tried to modify the model path and use other T5 model, but the model cannot be loaded? May I know how to train a model from s…
-
Is it possible to run a created model on an actual MetaTrader5 Instance?