microsoft / TransformerCompression

For releasing code related to compression methods for transformers, accompanying our publications
MIT License
354 stars 31 forks source link

How to load the tuned slicemodel #140

Closed qxpBlog closed 3 months ago

qxpBlog commented 4 months ago

@mtodd @tpope @sverrejoh @radical @ijbaird After I finished fine-tuning, the content of the folder I obtained was exactly the same as the content of the file after slicing, so where is the weight of fine-tuning and how should I load the fine-tuning model. image

nailimixaM commented 3 months ago

You can load the finetuned model in exactly the same way as you load the sliced model, see this method