Pierre Guillou

Hi, ONNX allows to compress transformers models and speed up the inference time on CPU and GPU. Who could share code / notebook to convert mT5 and ByT5 models to...

the method get_onnx_model() should not need the path to original model

2

Hi, After having obtained our ONNX model thanks to the `export_and_get_onnx_model()` method, we want to load it in order to use it in production. There is a method called `get_onnx_model()`...

## Environment info - `adapter-transformers` version: 2.0.1 - `DeepSpeed` version: 0.4.0 ## Details Hugging Face allows the use of [DeepSpeed](https://github.com/microsoft/DeepSpeed) in order to acelerate the training of a model in...

FastCorrect: are you planning to release the code?

1

Hi, I loved your paper on [FastCorrect](https://arxiv.org/pdf/2105.03842.pdf) (and FastCorrect 2) but I did not find any released code. I would like to apply FastCorrect on a Portuguese dataset. Are you...

Hi, I wanted to make inference with `LayoutXLM `with model parameters to `Half `(`float16`) dtype on CPU (I did try on GPU and it worked). As I'm using Transformers from...

Hi, I'm using Hugging Face libraries in order to run `LayoutXLM`. How can I decrease inference time? Which code to use? I've already tested `BetterTransformer `(`Optimum`) and `ONNX `but none...

How to change LayoutLMv3 to a LayoutLMv3 XLM (i.e. LayoutXLM-like model)

6

Hi @NielsRogge, I plan to finetune a LayoutXLM large like model. Why "like model"? Because until now, Microsoft did not release LayoutXLM large mas only a version base. As I...

Inference with nielsr/lilt-xlm-roberta-base

13

Hi @NielsRogge, in your notebook [Fine_tune_LiLT_on_a_custom_dataset%2C_in_any_language.ipynb](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/LiLT/Fine_tune_LiLT_on_a_custom_dataset%2C_in_any_language.ipynb), you do not write the code for inference from a document image. Can you give it please? I did try to copy the code...

How to decrease inference time of LiLT?

3

Hi, I'm using Hugging Face libraries in order to run `LiLT`. How can I decrease inference time? Which code to use? I've already try `BetterTransformer ` (`Optimum`) and `ONNX `...

Combining LiLT with XML RoBERTa large fails

2

Hi @NielsRogge, Many thanks for your notebook [Create_LiLT_+_XLM_RoBERTa_base.ipynb](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/LiLT/Create_LiLT_%2B_XLM_RoBERTa_base.ipynb). I try to use it for combining LiLT with [XML RoBERTa large](https://huggingface.co/xlm-roberta-large) but it failed. Here the code: ``` python gen_weight_roberta_like.py \...