facebookresearch / ConvNeXt-V2

Code release for ConvNeXt V2 model

facebookresearch/ConvNeXt-V2 Issues

cannot import name 'MinkowskiDepthwiseConvolution' from 'MinkowskiEngine'.
Closed a year ago5
Why not provide 22k-supervised finetuning model??? I am really shocked by that every available ConvNeXt-V2 pre-training weights has been finetuned on imagenet-1k. Please make 22k-supervised ConvNeXt-V2 open just like ConvNeXt-V1 !!!!!! 🙏🙏🙏
Updated 2 months ago
ImageNet 22k(21k) Traning loss at the end of training
Updated 3 months ago
Finetuning with limited Labels ?
Updated 3 months ago
[Question] Cosine Similarity
Updated 3 months ago
Onnx Export
Updated 3 months ago
Reason for double mask in Jax implementation?
Closed 4 months ago2
deployment issue in trt fp16
Updated 4 months ago
Pre-trained weights incompatible with backbone
Updated 4 months ago
could you share the dense masked conv based sparse encoding ?
Updated 4 months ago2
Weights trained using pre-training script produce NaN values when running fine-tuning script
Updated 5 months ago
Cannot install MinkowskiEngine with provided instructions
Updated 5 months ago1
Unable to update submodule (publickey error)
Updated 5 months ago4
Could you release a pre-trained model with a decoder?
Updated 6 months ago2
GRN can be used on any convent with FCMAE ？anyone tried this ?
Updated 6 months ago
Will the methods work on other CNNs?
Closed a year ago2
ImageNet-1K pre-trained weights ConvNeXt V2 supervised
Updated 8 months ago
visualization and feature consine distance
Updated 8 months ago2
Adapting ConvNetV2 for Time Series: Inconsistencies between Pre-training Visuals and Fine-Tuning Performance
Updated 9 months ago1
Make image and patch sizes dynamic
Updated 9 months ago
About the size of the patches
Updated 9 months ago8
cd MinkowskiEngine reported an error
Updated 10 months ago1
Could you please provide a demo code for how to visualize the effect of mask decoder, such as the following figure?
Updated 10 months ago2
On masking input images
Updated 10 months ago1
unable to install apex
Updated a year ago2
How can I reproduce the supervised training results?
Updated a year ago1
segmentation
Updated a year ago
The reconstruction images have a lot of noise
Closed a year ago5
Question on decoder resolution
Updated a year ago1
The official pre-training weights does not match the parameters of the model
Updated a year ago1
cd MinkowskiEngine报错，没有这个目录文件，是咋回事
Closed a year ago1
What is the loss of the pretrain model in fcmae stage?
Updated a year ago
Problem with ConvNeXt V2-B 384x384 weights
Closed a year ago4
How about the log files of pretraining?
Updated a year ago1
custom LayerNorm
Updated a year ago
No key named 'lr_scale' as input parameter to optimizer.AdamW
Closed a year ago
Training Setting for ConvNeXt V1-A Supervised (75.7%) in Table 14
Updated a year ago
RuntimeError: Error(s) in loading state_dict for ConvNeXtV2:
Updated a year ago5
MinkowskiLinear is missing from MinkowskiEngine, MinkowskiGELU code files
Closed a year ago
How to get the predicted results of the test set?
Updated a year ago
Do you consider comparing the scores of Cityscapes in the future to facilitate the data disclosure
Updated a year ago
Question regarding sparse convolution dimension
Updated a year ago
Reproduction of segmentation networks
Updated a year ago
LayerNorm vs torch.nn.LayerNorm
Updated a year ago1
The model used contains deepcopy, and a TypeError: cannot pickle 'MinkowskiConvolutionFunction' object error was reported when using this project
Updated a year ago
self-supervised pre-trained weights miss Parameters
Updated a year ago
parameter shape mismatch in GRN
Closed a year ago1
ImageNet-1K Benchmarks (Trained from scratch)
Closed a year ago1
Is the gamma and beta initial value wrong?Does zero work？
Closed a year ago
这个文件夹MinkowskiEngine是空的，没有setup文件
Closed a year ago