facebookresearch ConvNeXt issues

facebookresearch / ConvNeXt

Code release for ConvNeXt model

MIT License

5.78k stars 696 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

My model performs very very bad on NICO dataset

#99 huanranchen opened 2 years ago
1
Gradient disappears in CIFAR10

#98 huanranchen closed 2 years ago
1
Ques about LearningRateDecayOptimizer

#97 huixiancheng closed 2 years ago
2
return torch.full((x.size()[0], num_classes), off_value, device=device).scatter_(1, x, on_value) RuntimeError: CUDA error: device-side assert triggered

#96 cxq1 opened 2 years ago
0
About BN replace LN

#95 kings-rgb opened 2 years ago
1
The bug of resume/auto_resume

#94 anonymoussss closed 2 years ago
1
no mmcv_custom.load_checkpoint

#93 gaopeng91 opened 2 years ago
4
I can't find the referred dataset ImageNet-1K and ImageNet-22K

#92 DiegoIE opened 2 years ago
1
bugfix - cosine scheduler with warmup steps

#91 hmichaeli opened 2 years ago
0
Segmentation Error

#90 HuaqingHe opened 2 years ago
1
multiprocessing.errors

#89 Shiweiliuiiiiiii closed 2 years ago
1
What's the difference between model and model_ema?

#88 leviome closed 2 years ago
2
Not able to get 82.1 acc@1 for ConvNeXt-T (single machine training for ImageNet-1k)

#87 shantanuj closed 2 years ago
2
Multiple nodes training problem

#86 shawnye1994 opened 2 years ago
1
my model_ema not updat

#85 TAOSHss closed 2 years ago
9
Speed of the convnext

#83 shengkelong closed 2 years ago
5
ConvNext-base training speed slower than ViT-B

#82 exnx opened 2 years ago
1
'ConvNeXt is not in the models registry' . How to register the ConvNeXt ,i add the all files to Swin-T ，but did not work

#81 joeyslv closed 2 years ago
11
最后一个stage的LN在FP16推理时会溢出，导致输出结果错误

#80 shuaizzZ opened 2 years ago
4
Add backslash to command in TRAINING.md

#79 oikosohn opened 2 years ago
3
pretrained = None?

#78 Williamlizl closed 2 years ago
2
Invalid weights for isotropic ConvNeXt-L.

#77 sebastian-sz closed 2 years ago
2
ObjectDetection: Which MMdetection version of this object detection demo is used?

#76 chfguo closed 2 years ago
4
How to eval my model on custom data

#75 anonymoussss closed 2 years ago
0
convnext_tiny与r50在imagenet1k的准确率及flops对比，请问这是什么原因

#74 jackyyang-cv opened 2 years ago
2
size mismatch when evaluating

#73 esalagran closed 2 years ago
4
How to train ConvNeXt-tiny using an input size 384 ?

#72 K-H-Ismail closed 2 years ago
2
the model is not stable when convert to half

#71 JianyuanYin closed 2 years ago
1
ConvNeXt is added to HuggingFace Transformers

#70 gante opened 2 years ago
2
have you try convnext-t with mask rcnn 1x schedule? or can you provide the 3x log.

#69 czczup closed 2 years ago
12
how to use syncbn when training with more than 1 card?

#68 clw5180 closed 2 years ago
2
I add the configs and models of ConvNeXt to the Swin-Transformer-Object-Detection.

#67 Williamlizl closed 2 years ago
3
Typo in Fig.3 (b)

#66 youngwanLEE closed 2 years ago
1
您好，想问一下这段数据增强，如果我不想要的话，请问要怎么去写，因为好像它会跳转到timm.data,我不想要RandomResizedCropAndInterpolation这个操作，您好，感谢您的回答

#65 lxwkobe24 closed 2 years ago
1
Object Detection implementation issues

#64 cydiachen closed 2 years ago
3
u-net convnext

#63 seekingdeep closed 2 years ago
2
where are these files: ./_base_/models/cascade_mask_rcnn_convnext_fpn.py', '../_base_/datasets/coco_instance.py', '../_base_/schedules/schedule_1x.py',

#62 Williamlizl closed 2 years ago
9
The object detection data structure is the same as coco-format?

#61 Williamlizl closed 2 years ago
1
Results on ImageNet-C

#60 zhoudaquan closed 2 years ago
1
fix a typo in Install.md

#59 developer0hye opened 2 years ago
0
expected results on CIFAR

#58 hmichaeli closed 2 years ago
2
Hello I have an initial problem

#57 iamwic closed 2 years ago
12
Only object detection

#56 Protozzzer opened 2 years ago
2
Evaluation in own classes

#55 esalagran closed 2 years ago
3
ConvNeXt-S and ConvNeXt-T models pretrained on ImageNet-22k

#54 CanyonWind closed 2 years ago
6
Default image input size pre-trained COCO models?

#53 Michelvl92 closed 2 years ago
2
Fixed command line argument '--enable_wandb False' not actually disabling wandb.

#52 ericmintun closed 2 years ago
2
Meet unexcepted key when creating convnext_isotropic_large pretrained model

#51 rentainhe closed 2 years ago
4
Including a note on the TF/Keras models

#50 sayakpaul opened 2 years ago
0
COCO training: _base_/datasets/coco_instance.py" does not exist

#49 Michelvl92 closed 2 years ago
1

Previous Next