Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / kssteven418/i-bert issues and pull requests
#34 - error when runing download_glue_data.py
Issue -
State: open - Opened by amosrohe 7 months ago
Labels: bug
#34 - error when runing download_glue_data.py
Issue -
State: open - Opened by amosrohe 7 months ago
Labels: bug
#33 - python run.py --arch roberta_base --task STS-B
Issue -
State: open - Opened by Arthur-Ling 9 months ago
Labels: question
#33 - python run.py --arch roberta_base --task STS-B
Issue -
State: open - Opened by Arthur-Ling 9 months ago
Labels: question
#32 - About CUDA out of memory
Issue -
State: open - Opened by Ai-ZL 10 months ago
Labels: question
#32 - About CUDA out of memory
Issue -
State: open - Opened by Ai-ZL 10 months ago
Labels: question
#31 - Cannot perform quantization-aware-finetuning due to NaN values
Issue -
State: open - Opened by AmosHason about 2 years ago
Labels: bug
#30 - Fix NaN values in IntLayerNorm
Pull Request -
State: open - Opened by AmosHason about 2 years ago
- 1 comment
#29 - Task name references in strings are wrong
Issue -
State: open - Opened by AmosHason about 2 years ago
- 1 comment
Labels: bug
#28 - Fix STS-B name reference
Pull Request -
State: open - Opened by AmosHason about 2 years ago
- 1 comment
#28 - Fix STS-B name reference
Pull Request -
State: open - Opened by AmosHason about 2 years ago
- 1 comment
#27 - Fix task name references
Pull Request -
State: open - Opened by AmosHason about 2 years ago
- 1 comment
#27 - Fix task name references
Pull Request -
State: open - Opened by AmosHason about 2 years ago
- 1 comment
#26 - Where can I find the integer-sqrt kernel ?
Issue -
State: open - Opened by Alex-Songs over 2 years ago
Labels: question
#26 - Where can I find the integer-sqrt kernel ?
Issue -
State: open - Opened by Alex-Songs over 2 years ago
Labels: question
#25 - 0
Issue -
State: open - Opened by Arouniversal over 2 years ago
Labels: bug
#24 - About scaling_factor
Issue -
State: open - Opened by DOHA-HWANG over 2 years ago
Labels: question
#24 - About scaling_factor
Issue -
State: open - Opened by DOHA-HWANG over 2 years ago
Labels: question
#23 - Pre-trained weights for specific tasks
Issue -
State: open - Opened by roymiles over 2 years ago
Labels: question
#23 - Pre-trained weights for specific tasks
Issue -
State: open - Opened by roymiles over 2 years ago
Labels: question
#22 - Storing both float32 and int parameters
Issue -
State: open - Opened by huu4ontocord almost 3 years ago
Labels: bug
#22 - Storing both float32 and int parameters
Issue -
State: open - Opened by huu4ontocord almost 3 years ago
Labels: bug
#21 - Latency 20x with quant_mode = true
Issue -
State: open - Opened by LiamPKU almost 3 years ago
- 1 comment
Labels: question
#21 - Latency 20x with quant_mode = true
Issue -
State: open - Opened by LiamPKU almost 3 years ago
- 1 comment
Labels: question
#20 - Arguments in run.py
Issue -
State: open - Opened by ZhangYunchenY almost 3 years ago
Labels: bug
#20 - Arguments in run.py
Issue -
State: open - Opened by ZhangYunchenY almost 3 years ago
Labels: bug
#19 - Wrong script of downloading GLUE datasets
Issue -
State: open - Opened by zzf1130 almost 3 years ago
Labels: bug
#19 - Wrong script of downloading GLUE datasets
Issue -
State: open - Opened by zzf1130 almost 3 years ago
Labels: bug
#18 - IBert problems of quant_model=true
Issue -
State: open - Opened by CSuperlei over 3 years ago
- 1 comment
Labels: bug
#17 - Another setting for quantization
Issue -
State: open - Opened by gksruf over 3 years ago
Labels: question
#17 - Another setting for quantization
Issue -
State: open - Opened by gksruf over 3 years ago
Labels: question
#16 - How can we change the quantization settings?
Issue -
State: open - Opened by kentaroy47 over 3 years ago
Labels: question
#16 - How can we change the quantization settings?
Issue -
State: open - Opened by kentaroy47 over 3 years ago
Labels: question
#15 - Can not inference the quantilized model in my device by int8
Issue -
State: open - Opened by deepfind over 3 years ago
- 1 comment
#15 - Can not inference the quantilized model in my device by int8
Issue -
State: open - Opened by deepfind over 3 years ago
- 1 comment
#14 - why is Integer-only finetuning is much more slower than fp32 finetune
Issue -
State: closed - Opened by renmada over 3 years ago
Labels: question
#14 - why is Integer-only finetuning is much more slower than fp32 finetune
Issue -
State: closed - Opened by renmada over 3 years ago
Labels: question
#13 - Is ibert-roberta-base on huggingface model hub the same as roberta-base
Issue -
State: closed - Opened by renmada over 3 years ago
Labels: question
#12 - How is the scaling factor S implemented with integer?
Issue -
State: open - Opened by deJQK over 3 years ago
Labels: question
#11 - 3rd order polynomial approximation to GeLU
Issue -
State: closed - Opened by CaoZhongZ over 3 years ago
Labels: question
#10 - Bugs in the code
Issue -
State: open - Opened by hsiehjackson over 3 years ago
Labels: bug
#10 - Bugs in the code
Issue -
State: open - Opened by hsiehjackson over 3 years ago
Labels: bug
#9 - rationale considering in using floor or round
Issue -
State: open - Opened by CaoZhongZ over 3 years ago
- 1 comment
Labels: question
#8 - (huggingface) The output of IBERT is float. Am I doing wrong?
Issue -
State: closed - Opened by kyoungrok0517 almost 4 years ago
Labels: question
#7 - quantize other roberta model
Issue -
State: open - Opened by longyueling almost 4 years ago
Labels: question
#6 - Quantization on trained model
Issue -
State: open - Opened by shon-otmazgin almost 4 years ago
- 10 comments
Labels: question
#5 - Why use 22 bit quantized activations for some layer norms (except in Embeddings)?
Issue -
State: closed - Opened by bdalal almost 4 years ago
- 2 comments
Labels: question
#4 - Possible bug in IntSoftmax
Issue -
State: open - Opened by bdalal almost 4 years ago
Labels: bug
#3 - Training in mixed precision
Issue -
State: closed - Opened by bdalal almost 4 years ago
- 3 comments
Labels: question
#2 - Missing deployment part on TensorRt
Issue -
State: open - Opened by pommedeterresautee almost 4 years ago
- 3 comments
Labels: question
#1 - Can use the CPU in the inference state?
Issue -
State: open - Opened by luoling1993 almost 4 years ago
- 1 comment
Labels: question