microsoft/DeBERTa issues and pull requests

#54 - Question about the hyperparameters of SuperGlue

Issue - State: open - Opened by oyxuan-11 over 3 years ago

#53 - DeBERTa base different performance numbers

Issue - State: closed - Opened by DarshanPatel11 over 3 years ago - 1 comment

#52 - where is the absolute position embeddings?

Issue - State: open - Opened by ylwangy over 3 years ago

#51 - is this a bug? in disentangled_attention.py pos_query_layer's dimension is 3, when use p2p attention and this code:\n pos_query = pos_query_layer[:,:,att_span:,:] \n get IndexError: too many indices for tensor of dimension 3

Issue - State: open - Opened by hj-github1256 over 3 years ago

#50 - missing import of unicodedata package

Pull Request - State: closed - Opened by alisafaya over 3 years ago

#49 - How does it save computational cost in EMD?

Issue - State: open - Opened by zhouxincheng over 3 years ago

#48 - Penhe/merge ms

Pull Request - State: closed - Opened by BigBird01 over 3 years ago

#47 - A question about the implementation of position-to-content attention

Issue - State: closed - Opened by SparkJiao over 3 years ago

#46 - The exact English pretraining data and Chinese pretraining data that are exact same to the BERT paper's pretraining data.

Issue - State: open - Opened by guotong1988 over 3 years ago

#45 - The amount of WSC training examples

Issue - State: closed - Opened by slowwavesleep over 3 years ago - 1 comment

#44 - A question about the Attention map

Issue - State: closed - Opened by 14H034160212 over 3 years ago - 1 comment

#43 - [ONNX] Add symbolic function so XSoftmax can be exported to ONNX.

Pull Request - State: closed - Opened by fatcat-z over 3 years ago - 1 comment

#42 - can't load v1 model

Issue - State: open - Opened by li1117heex over 3 years ago - 2 comments

#41 - Enhanced Masked Decoding

Issue - State: closed - Opened by joaogui1 over 3 years ago - 1 comment

#40 - model key 'encoder.layer.0.attention.self.query_proj.weight' not found in base-mnli

Issue - State: open - Opened by cedar33 almost 4 years ago - 1 comment

#39 - "deberta-v2-xxlarge"-Model not working!

Issue - State: closed - Opened by kinimod23 almost 4 years ago - 2 comments

#36 - Speed of DeBERTa seems disappointing

Issue - State: closed - Opened by shenfe almost 4 years ago - 1 comment

#33 - RuntimeError: Index tensor must have the same number of dimensions as input tensor

Issue - State: open - Opened by lgstd almost 4 years ago - 5 comments

#28 - Is SiFT included?

Issue - State: closed - Opened by applenob almost 4 years ago - 1 comment

#22 - Is there Chinese version？

Issue - State: closed - Opened by MrRace almost 4 years ago - 2 comments

#14 - HTTP Error 403: Forbidden when downloading glue_tasks

Issue - State: closed - Opened by huberemanuel almost 4 years ago - 4 comments

#11 - Does any plan to release the pretrain code?

Issue - State: closed - Opened by RyanHuangNLP about 4 years ago - 4 comments

#6 - [WIP] ONNX conversion

Pull Request - State: open - Opened by ganik over 4 years ago - 1 comment

Ecosyste.ms: Issues

GitHub / microsoft/DeBERTa issues and pull requests