yoshitomo-matsubara/torchdistill
A coding-free framework built on PyTorch for reproducible deep learning studies. 🏆25 knowledge distillation methods presented at CVPR, ICLR, ECCV, NeurIPS, ICCV, etc are implemented so far. 🎁 Trained models, training logs and configurations are available for ensuring the reproducibiliy and benchmark.
PythonMIT
Issues
- 6
get an error
#481 opened by cxchen100 - 1
Hi @cxchen100 ,
#482 opened by cxchen100 - 4
Possible re-implementation of KD w/ LS
#478 opened by sunshangquan - 1
Knowledge distillation related
#476 opened by andynnnnn - 2
[BUG] Problems in DistillationBox
#471 opened by 1396066796 - 2
- 2
Incorrect conditional judgement
#467 opened by 1396066796 - 1
- 3
- 1
is tochdistill support knowlede distillation for Vision Foundation Models like Grounding Dino / Grounding DinoSAM ?
#427 opened by solomonmanuelraj - 2
[BUG]ImportError: cannot import name 'import_dependencies' from 'torchdistill.common.main_util'
#398 opened by zhangruju - 1
[BUG] Missing Link in Readme
#389 opened by m-parchami - 4
[BUG] fp16 causes AssertionError: No inf checks were recorded for this optimizer
#386 opened by jsrdcht - 1
I tried with this script also, only single nproc seems to be working. Do i need to define any additional enviornment variables like RANK or LocaL HOST
#379 opened by nighting0le01 - 1
[BUG] Not supported to Nvidia 4090
#367 opened by allent4n - 1
How should I use Torchdistill?
#365 opened by 2842193395 - 1
Not a bug but a discrepency between the log and config file for kd-resnet18_from_resnet34
#279 opened by Coderx7 - 1
Where is trained model?
#259 opened by Holmes2002 - 1
Custom Data
#255 opened by Holmes2002 - 1
Use different models as Teacher/Student
#246 opened by jaideep11061982 - 1
Disagreement betweeen the log and configuration of kd-resnet18_from_resnet34
#237 opened by Calmepro777 - 1
- 1
Why using `log_softmax` instead of `softmax`?
#233 opened by nguyenvulong - 1
How to train my own COCO dataset for object detection?
#222 opened by Muke6 - 2
Similarity Preserving KD
#221 opened by hanoonaR - 3
- 4
- 9
Combine two distillation losses
#215 opened by shwinshaker - 24
- 2
If the Teacher model is different from Student model, how can I use this framework?
#133 opened by topbookcc - 2
Bug. Bad implement.
#214 opened by aiboys - 3
Distilling Knowledge from a image classification model with sigmoid function and binary cross entropy
#211 opened by publioelon - 1
It seems some bug in `split_dataset`
#209 opened by aiboys - 2
Affinity Loss usage
#181 opened by ChidanandKumarKS - 5
- 3
Hyperparameters tunning
#154 opened by AhmedHussKhalifa - 2
Using forward hook for auxiliary loss
#152 opened by 1ho0jin1 - 7
Implementation of SemCKD
#140 opened by AndyFrancesco29 - 3
Support for SSD Object Detection Model?
#117 opened by anujdutt9 - 15
- 2
CSE-L2 KD mobilenetv2 from resnet18 on cifar100
#131 opened by wongyufei - 1
How to specify the weight of the downloaded teacher model without access to the Internet?
#121 opened by wongyufei - 2
AssertionError: DistributedDataParallel is not needed when a module doesn't have any parameter that requires a gradient.
#122 opened by wongyufei - 5
Segmentation fault encountered when entering the second epoch with num_workers>0
#90 opened by RulinShao - 6
ForwardHookManager on multiple GPUs
#32 opened by arbellea - 8
- 1
About dockerfile of torchdistill
#36 opened by lliai - 2