facebookresearch / ConvNeXt

Code release for ConvNeXt model

facebookresearch/ConvNeXt Issues

Resize Test Images for size less than 32
Updated 7 months ago
How can I make predictions on my testing dataset for using command line?
Closed 8 months ago
I can not use 'timm.create_model' to create my convnext
Updated 9 months ago1
Error:0.3.2timm.create_model did not have layer_scale_init_value
Updated 10 months ago
convnext_tiny Loss not change
Updated 10 months ago1
A warning appears in DDP using ConvNeXt
Updated 10 months ago1
cannot load all 22k_models
Updated 10 months ago2
Missing code or wrong reference
Updated 10 months ago2
Hyperparameter setting for training from scratch on CIFAR-10
Updated 10 months ago9
Finetuning on Imagenet100 using ConvNeXt-T for ImageNet1k
Closed a year ago
load pretrained model, there are errors. RuntimeError: Error(s) in loading state_dict for ConvNeXt: Missing key(s) in state_dict
Updated a year ago1
Infinite loss value when training under amp
Updated a year ago1
Welcome update to OpenMMLab 2.0
Updated a year ago
input dimensions
Updated a year ago
Tensorrt Model?
Updated a year ago
LayerNorm, what is going on?
Updated a year ago1
can convNext add deformable conv in the block
Updated a year ago
Number Of Classes
Closed a year ago2
Error in Convnext training with Swin transformer framework - The testing results of the whole dataset is empty.
Updated a year ago
Why LayerNorm before conv in downsampling layers ?
Updated a year ago
Large Variance in feature maps
Updated a year ago8
Ask for release model on COCO
Updated a year ago
The model and loaded state dict do not match exactly when trying to load pretrained weights
Updated a year ago2
head layer can't be replaced in convnex like other(classifier layer) pytorch model?
Updated 2 years ago1
Function of drop_path——drop_path应该设置多少，模型能提高map喃
Closed 2 years ago4
Channel Last memory layout
Updated 2 years ago
inference
Closed 2 years ago
convnext load pretrained model, but backbone.norm has no weight
Updated 2 years ago
there is no need to rewrite the 'class LayerNorm(nn.Module)'
Updated 2 years ago5
ValueError: loaded state dict has a different number of parameter groups
Updated 2 years ago2
Can't run the model using RTX3080
Closed 2 years ago
Use convnext's training settings to train resnet
Updated 2 years ago
What is the difference between "LearningRateDecayOptimizerConstructor" with the default one?
Updated 2 years ago1
Kindly asking for all variants/upgrades mentioned in the paper
Updated 2 years ago
Ade20k dataset is not available
Closed 2 years ago1
why use the class LayerNorm()? why not using the nn.LayerNorm(dim) in pytorch?
Closed 2 years ago1
model_ema
Closed 2 years ago1
Error in importing backbones from mmdet in object detection module.
Updated 2 years ago
Error on Validation
Updated 2 years ago
__init__() got an unexpected keyword argument 'pretrained'
Updated 2 years ago4
Trying to train ConvNeXt-S but it doesn't seem to work
Closed 2 years ago
APs and AP-large detail in detection?
Updated 2 years ago
Could you provide the example of semantic segmentation in detail for training custom datasets from scratch?
Updated 2 years ago
Why do you need global sync?
Updated 2 years ago
Training logs for different seeds and variants
Updated 2 years ago
Ques about LearningRateDecayOptimizer
Closed 2 years ago2
What version of Pillow should I use?
Updated 2 years ago
My model performs very very bad on NICO dataset
Updated 2 years ago1
Gradient disappears in CIFAR10
Closed 2 years ago1
return torch.full((x.size()[0], num_classes), off_value, device=device).scatter_(1, x, on_value) RuntimeError: CUDA error: device-side assert triggered
Updated 2 years ago