Issue when loading model checkpoints

[hieudx149]

Hi there, When i load your model checkpoint base or xsmall, i see this warning `Some weights of the model checkpoint at Fsoft-AIC/videberta-base were not used when initializing DebertaV2Model: ['mask_predictions.LayerNorm.weight', 'deberta.embeddings.word_embeddings._weight', 'mask_predictions.LayerNorm.bias', 'mask_predictions.dense.weight', 'mask_predictions.classifier.weight', 'mask_predictions.classifier.bias', 'mask_predictions.dense.bias']

• This IS expected if you are initializing DebertaV2Model from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).
• This IS NOT expected if you are initializing DebertaV2Model from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). ` it's seem to missing embeddings layer checkpoint when loading model !

[phatjkk]

I have the same problem too. Can you tell me how to fix it? Thanks.

[anhquan0412]

Same questions as well. Looks like there's no embedding weight (both base and xsmall) and when I used this for my fine-tuning task, it seems that the model was unable to learn,