lukemelas/EfficientNet-PyTorch issues and pull requests

#348 - Does EfficientNet have pre-trained weights in ImageNet1k?

Issue - State: open - Opened by apuomline 4 months ago - 1 comment

#347 - Get the number of output features from the last layer and remove the last layer?

Issue - State: open - Opened by srevandro 6 months ago

#346 - Are you planning to share eff autoencoders?

Issue - State: open - Opened by osivaz61 8 months ago

#345 - Very low validation accuracy on training from scratch on CIFAR10

Issue - State: open - Opened by appledora 8 months ago

#344 - Why is the depth of efficientnet-b7 defined as 3.1?

Issue - State: open - Opened by T990820 12 months ago

#343 - How to add additional layer( inside the features module) in pre-trained EfficientNetB0 model?

Issue - State: open - Opened by sashika-himali about 1 year ago - 2 comments

#342 - How to train a new model on the imagenet dataset？

Issue - State: open - Opened by figurative666 over 1 year ago

#341 - RuntimeError: ONNX export failed: Couldn't export Python operator SwishFunction

Issue - State: open - Opened by DAVID-Hown over 1 year ago

#340 - What is your training strategy for the pretrained models ?

Issue - State: open - Opened by JGyoung-UCAS over 1 year ago

#339 - multilabel multiclass image classification

Issue - State: open - Opened by crabmon almost 2 years ago

#338 - What's the performance of the models trained with adaversial training?

Issue - State: open - Opened by Zhaoyi-Yan almost 2 years ago

#337 - how to use the include_top parameter in pretrained model?

Issue - State: open - Opened by lck666666 about 2 years ago

#336 - Can not script the model with torchscript.

Issue - State: open - Opened by PhanThanhTrung about 2 years ago

#335 - How to train

Issue - State: open - Opened by li-an-sheng about 2 years ago - 2 comments

#334 - EfficientNetV2-PyTorch

Issue - State: open - Opened by LB15642219539 about 2 years ago - 1 comment

#333 - EfficientNetV2-PyTorch

Issue - State: closed - Opened by LB15642219539 about 2 years ago

#332 - Changing to num class to train to smaller than default

Issue - State: open - Opened by lchunleo over 2 years ago

#331 - Add type hints

Issue - State: open - Opened by pietermarsman over 2 years ago

#330 - Pretrained _conv_stem.weight different each run when in_channels=1

Issue - State: open - Opened by daniel-a-diaz over 2 years ago

#329 - efficient

Issue - State: open - Opened by the-cat-crying over 2 years ago

#328 - from_pretrained() got an unexpected keyword argument 'weights_path'

Issue - State: open - Opened by onlinehuazai over 2 years ago - 1 comment

#327 - What data set is the efficientnet pretraining model based on？imagenet-1k or imagenet-21k

Issue - State: open - Opened by ImportAI over 2 years ago

#326 - Transfer learning on pretrained models

Issue - State: closed - Opened by chinembedded over 2 years ago - 2 comments

#325 - Allow `verbose` arg through `from_pretrained`

Pull Request - State: closed - Opened by robin-wayve over 2 years ago

#324 - got it

Issue - State: closed - Opened by fire717 over 2 years ago

#323 - dropout & drop_connect

Issue - State: closed - Opened by M-junjun over 2 years ago - 2 comments

#322 - It doesn't work when I set the in_channels=1 using from_name to define an efficientnet-b5.

Issue - State: open - Opened by shijun18 over 2 years ago

#321 - Training details for pretrained models

Issue - State: open - Opened by JGyoung33 over 2 years ago

#320 - References for efficientnet-b8

Issue - State: open - Opened by mliwang over 2 years ago

#319 - the pytorch vision needed 1.10.2

Issue - State: closed - Opened by myh12138 over 2 years ago

#318 - How can I solve this probrem

Issue - State: closed - Opened by lry-123456789 almost 3 years ago - 1 comment

#317 - The parameters and FLOPs do not match the paper

Issue - State: open - Opened by cj-wade almost 3 years ago - 1 comment

#316 - Extract features and endpoints in one pass

Issue - State: open - Opened by jsetty almost 3 years ago

#315 - Error when converting tf checkpoint to pytorch weights

Issue - State: open - Opened by alancarlosml almost 3 years ago

#314 - Issues while saving a TorchScript Model(Anynet) using torch.jit.save()

Issue - State: open - Opened by nakul3112 almost 3 years ago

#313 - extract till Conv2dStaticSamePadding: 3-245 in efficientnet

Issue - State: open - Opened by Vivek8080kumar almost 3 years ago

#312 - How to train the model from scratch in a different dataset

Issue - State: open - Opened by Agito555 almost 3 years ago

#311 - Using Colab version of efficientNet on State Farm distracted driver dataset

Issue - State: open - Opened by alexcoulter15 almost 3 years ago

#310 - removed a pitfall where a built-in name was being shadowed

Pull Request - State: open - Opened by NaelsonDouglas about 3 years ago

#309 - EfficientNet V2

Issue - State: open - Opened by alancarlosml about 3 years ago - 2 comments

#308 - model predict error

Issue - State: open - Opened by stringk245 about 3 years ago

#307 - what should I change to be able to use a data input of 720*184 ?

Issue - State: open - Opened by Oussamab21 about 3 years ago

#306 - what does line 'pip install -e .' do?

Issue - State: open - Opened by shaoshuaidu about 3 years ago

#305 - What is the function "extract_endpoints"?

Issue - State: open - Opened by PeterKim1 about 3 years ago

#304 - Test with single image gives different result in each runtime

Issue - State: open - Opened by gihunsong about 3 years ago - 9 comments

#303 - Waiting for new efficientnetv2!🥳

Issue - State: open - Opened by LovelyQuantum about 3 years ago

#302 - Error when I load the onnx model to the tensorrt format

Issue - State: open - Opened by ShunLu91 about 3 years ago - 2 comments

#301 - Got the Top-1 accu of only 70.7% when training efficientnet-b3 with Imagenet

Issue - State: open - Opened by YvetteLaw about 3 years ago - 3 comments

#300 - multi-head efficientNet

Issue - State: open - Opened by www7890 about 3 years ago

#299 - Update utils.py

Pull Request - State: open - Opened by sathyapriyaa-sketch over 3 years ago - 1 comment

#298 - Hello I'm curious about the set_swish (memory_efficient=False) used in onnx.export

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

#297 - Tried to implement Fuse-MBConv block, have not tested

Pull Request - State: open - Opened by vegardkoder over 3 years ago

#296 - Cannot export EfficientNet to onnx

Issue - State: closed - Opened by LamyaMohaned over 3 years ago - 2 comments

#295 - convert keras efficientnetb3 to pytorch model version? keras version 2 tf version?

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

#294 - the GPU program failed to execute at /opt/conda/conda-bld/pytorch_1556653215914/work/aten/src/THC/THCBlas.cu:259

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

#293 - Add input at intermediate layer

Issue - State: open - Opened by FrancescoMandru over 3 years ago - 1 comment

#292 - TensorRT implementation

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

#291 - half / mixed precision?

Issue - State: open - Opened by jacekpoplawski over 3 years ago - 1 comment

#290 - allow to load pretrained weights when the include_top variable is False

Pull Request - State: open - Opened by kurnianggoro over 3 years ago - 1 comment

#289 - Consider if user has GPU/CPU while calling `torch.load()`

Issue - State: open - Opened by BRO3886 over 3 years ago - 3 comments

#288 - A Doubt on segmentation

Issue - State: closed - Opened by Rituraj-commits over 3 years ago - 3 comments

#287 - An doubt about Conv2dDynamicSamePadding class

Issue - State: open - Opened by rose-jinyang over 3 years ago

#286 - How to compute the params?

Issue - State: open - Opened by linlll over 3 years ago - 1 comment

#285 - Efficient net is sensitve to interpolation

Issue - State: open - Opened by NingNanXin over 3 years ago - 1 comment

#284 - Gradient Checkpointing

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

#283 - Bug for multitask learning and all the fc layer replacement situations

Issue - State: open - Opened by XingjChen over 3 years ago - 1 comment

#282 - Mapping width_coefficient, depth_coefficient and image_size to paper's parameters

Issue - State: closed - Opened by fpaissan over 3 years ago - 2 comments

#281 - KeyError: 'efficientnet-b8'

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

#280 - AutoAugment Question

Issue - State: closed - Opened by RaivoKoot over 3 years ago - 3 comments

#279 - Is there the example code to train it with COCO?

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

#278 - Missing keys when loading pretrained weights

Issue - State: open - Opened by BavarianToolbox over 3 years ago - 7 comments

#273 - Training Speed is Slow

Issue - State: closed - Opened by ajaykrishnan23 over 3 years ago - 2 comments

#271 - _fc.weight error when loading model for transfer learning

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

#262 - Models trained in v0.6.3 give wrong output in v0.7.0

Issue - State: open - Opened by jamt9000 almost 4 years ago - 3 comments

#257 - AutoEncoder using the EfficientNet

Pull Request - State: open - Opened by xingyaoww almost 4 years ago - 5 comments

#241 - EfficientNet not avaialable in efficientnet_pytorch

Issue - State: closed - Opened by rnjbdya about 4 years ago - 1 comment

#231 - Pretrained weights don't give same accuracy as stated in repo

Issue - State: open - Opened by neiladit about 4 years ago - 6 comments

#219 - Getting High Loss Values

Issue - State: open - Opened by IamSparky over 4 years ago - 6 comments

#213 - from_name and num_classes

Issue - State: open - Opened by jacekpoplawski over 4 years ago - 4 comments

#208 - #192 include top

Pull Request - State: closed - Opened by nwschurink over 4 years ago - 4 comments

#204 - EfficientB0 on cifar10 without transfer learning and trainng from scratch

Issue - State: closed - Opened by ww-zwj over 4 years ago - 2 comments

#202 - Anyone achieved about 77% on imagenet with eff-b0??

Issue - State: closed - Opened by ooodragon94 over 4 years ago - 3 comments

#197 - How to change BN to GN and FRN layer in efficientnet?

Issue - State: open - Opened by mobassir94 over 4 years ago - 1 comment

#196 - which part has been changed to speed the inference?

Issue - State: closed - Opened by kaixin-bai over 4 years ago - 3 comments

#195 - EfficientNet-b0 slower than Resnet-50?

Issue - State: closed - Opened by cgebbe over 4 years ago - 3 comments

#189 - Transfer Learning not working

Issue - State: open - Opened by gost-sniper over 4 years ago - 12 comments

#185 - output of extract_feature is a nan tensor when eval model

Issue - State: closed - Opened by Kyeongpil over 4 years ago - 6 comments

#184 - Issues with applying GradCam on this implementation

Issue - State: open - Opened by MrGranddy over 4 years ago - 4 comments

#147 - out of memory

Issue - State: open - Opened by wangmingrun1314 over 4 years ago - 11 comments

#142 - transfer learning using efficientnet NotImplementedError

Issue - State: open - Opened by victorzhucompass over 4 years ago - 9 comments

#129 - Unable to load ONNX model into TensorRT

Issue - State: open - Opened by kaishian almost 5 years ago - 3 comments

#121 - What are the training settings on CIFAR-100?

Issue - State: closed - Opened by ChawDoe almost 5 years ago - 5 comments

#120 - Resume Training

Issue - State: closed - Opened by etetteh almost 5 years ago - 1 comment

GitHub / lukemelas/EfficientNet-PyTorch issues and pull requests