huggingface nn_pruning issues

huggingface / nn_pruning

Prune a model while finetuning or training.

Apache License 2.0

394 stars 58 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Is it possible to prune only particular (not all parameters) of the transformer ?

#43 mayank64ce opened 2 months ago
0
ValueError: mutable default <class 'nn_pruning.modules.masked_nn.InitDirective'> for field mask_init is not allowed: use default_factory

#42 xihajun closed 1 year ago
0
Does this script run with the hybrid-filled method by default?Thanks!

#41 iMountTai opened 1 year ago
1
Applying Block Movement Pruning for BART

#40 apurvnagvenkar opened 2 years ago
1
Quantization tests fail

#39 LukasHedegaard opened 2 years ago
1
Fix quantization tests

#38 LukasHedegaard opened 2 years ago
0
Is this pruning methods common for multiHeadAttention

#37 shoveller86 opened 2 years ago
0
nn_pruning doesn't seem to work for T5 Models, Roberta-based Models

#36 ghost opened 2 years ago
3
AttributeError: 'GraphModule' object has no attribute 'config'

#35 LYF915 opened 2 years ago
0
unable to replicate results of example notebooks

#34 rohanshingade opened 2 years ago
0
What is the difference between "finetune" and "final-finetune" in `/example`.

#33 eric8607242 opened 2 years ago
1
a bug for MNLI eval result?

#32 xwuShirley opened 2 years ago
1
Fix build issues

#31 deponovo closed 2 years ago
1
Is edge-popup the same as movement pruning with frozen weights?

#30 mariomeissner opened 2 years ago
0
Support for `BertModel`, `BartModel` (`AutoModel`)

#29 kainoj opened 3 years ago
0
Experiments run slowly on V100

#28 xiamengzhou closed 3 years ago
2
Not seeing the inference speed up on cuda using the sparse trainer notebook

#27 HamidShojanazeri opened 3 years ago
1
Is global sort supported?

#26 xiaowu0162 opened 3 years ago
0
Does this support TensorFlow?

#25 Melkeydev opened 3 years ago
0
Example links not working in docs/HOWTO

#24 gitlost-murali opened 3 years ago
0
Why you say it's not needed to run the models pruned by the nn_pruning tools？

#23 luofuli opened 3 years ago
1
Graph Mode QAT / Static Quantization

#22 michaelbenayoun closed 3 years ago
1
Adding summarization and translation.

#21 echarlaix closed 3 years ago
0
NN_pruning module for Question Answering

#20 fdlci closed 3 years ago
1
Adding T5 structure and fixing context module naming for encoder decoder architecture

#19 echarlaix closed 3 years ago
0
Modification to make SparseXP structure independent.

#18 echarlaix closed 3 years ago
0
Modifications to make inference model patcher structure independent.

#17 echarlaix closed 3 years ago
0
Integrating quantization aware training to SparseXP.

#16 michaelbenayoun closed 3 years ago
1
Adding BART architecture

#15 echarlaix closed 3 years ago
2
Added option to keep GeLU2ReLU transition parameters during evaluation

#14 echarlaix closed 3 years ago
0
Layer2NoNorm uses same mix and delta value during eval mode as during training (useful for debugging or analysis of Layer2NoNorm transition)

#13 echarlaix closed 3 years ago
1
Added squadv2 task

#12 echarlaix closed 3 years ago
1
Replaced task_name in dataset_name for text classification for homogeneity

#11 echarlaix closed 3 years ago
1
added several tasks of the GLUE benchmark

#10 echarlaix closed 3 years ago
1
Descriptions of hybrid pruning

#9 xiamengzhou opened 3 years ago
1
[WIP] Fix unit tests

#8 lewtun closed 3 years ago
2
Fix typo in dependency

#7 lewtun closed 3 years ago
0
Add text classification notebook

#6 lewtun closed 3 years ago
3
No weights removed during fine-pruning?

#5 lewtun closed 3 years ago
3
Dependencies and version numbers?

#4 lewtun closed 3 years ago
3
General help for custom token classification endtask

#3 JulesBelveze opened 3 years ago
6
Possible speed up for Sparse Block Computation

#2 tchaton closed 3 years ago
1
Fix readme table

#1 thomwolf closed 3 years ago
1