tianyic/only_train_once
OTOv1-v3, NeurIPS, ICLR, TMLR, DNN Training, Compression, Structured Pruning, Erasing Operators, CNN, Diffusion, LLM
PythonMIT
Issues
- 19
Models with no ZIG?
#57 opened by iamanigeeit - 2
- 2
Prune vision transformer
#60 opened by Nofear18 - 2
Can not prune YOLOv8 model
#59 opened by hoangtv2000 - 2
HESSO optimizer technical report
#58 opened by jiao1234567 - 1
Allowing dummy_input to be tuple / dict
#56 opened by iamanigeeit - 5
- 5
Error in Concat-Split
#50 opened by C0NGTRI123 - 1
Understanding pruning_steps
#55 opened by iamanigeeit - 2
pruning transformer-based model
#54 opened by huanghmingyue - 3
Confused about Fig 3a in OTO v1 paper
#52 opened by iamanigeeit - 2
When quantizing the YOLOv8 model and calling the code optimizer = oto.dhspg(), an error is thrown.
#38 opened by Gooddz1 - 1
- 1
How to load compressed model ?
#51 opened by WYYAHYT - 13
Grouped Conv compress ERROR
#39 opened by Eternal-Br - 6
Error when testing yolov5
#45 opened by jiao1234567 - 3
When I replaced the model in the tutorial with my own other model, this error occurred
#46 opened by zxf1336899252 - 9
Wrong train use DDP
#44 opened by C0NGTRI123 - 2
Problem with compress model
#32 opened by bnomq - 2
oto.compress() failure
#35 opened by chamecall - 1
How to set epsilon param of dhspg?
#27 opened by gaoxueyi0532 - 1
Get pytorch model prune shape
#42 opened by C0NGTRI123 - 1
- 2
- 7
- 1
I want to know if RNN(LSTM or GRU) is supported
#31 opened by AaZz101 - 3
- 1
- 4
two problems about algorithm convergence
#28 opened by gaoxueyi0532 - 1
New tutorials problem resnet50_imagenet.ipynb
#24 opened by Miocio-nora - 1
- 1
some confuse about Algorithm-3
#25 opened by gaoxueyi0532 - 1
google-protobuf depended by onnx is v2.6.1,but it conflicts with python3,maybe only python2 works well!
#22 opened by gaoxueyi0532 - 7
setting on imagenet
#18 opened by mountain111 - 2
Protocol Buffer Version Conflict with local
#20 opened by gaoxueyi0532 - 47
oto.compress failed with "xs.append(param.data.view(cc.num_groups, -1))" in graphy.py
#5 opened by songkq - 3
Problems when using OTOv2 on Yolov6
#17 opened by Miocio-nora - 1
OTO on detection models
#16 opened by LishaTang - 7
- 5
The compressed onnx model can't run inference.
#13 opened by rotorliu - 7
- 2
- 3
How to resume?
#9 opened by rotorliu - 1
ONNX opset setting
#7 opened by unique-null - 2
Interested in OTOV2 work
#4 opened by qgh1223