Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / facebookresearch/ConvNeXt-V2 issues and pull requests

#83 - where do I install APEX

Issue - State: closed - Opened by Mayolov about 1 month ago

#82 - Why GRN layer called a Normalization？

Issue - State: open - Opened by SnifferCaptain about 1 month ago

#81 - No decoder in released model checkpoints

Issue - State: open - Opened by Luke-Byrne-Eng 3 months ago - 1 comment

#76 - GRN in paper uses x_i / sum x_j whereas code uses x_i / mean x_j

Issue - State: open - Opened by sebLopezCot 7 months ago - 1 comment

#75 - Colab Implementation

Issue - State: open - Opened by emrekeles-arch 8 months ago

#74 - downsample_layer model order

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

#74 - downsample_layer model order

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

#73 - can grn add to resnet

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

#73 - can grn add to resnet

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

#70 - Finetuning with limited Labels ?

Issue - State: open - Opened by rohit-gupta 11 months ago

#69 - [Question] Cosine Similarity

Issue - State: open - Opened by dhkim0225 11 months ago

#68 - Onnx Export

Issue - State: open - Opened by sulaimanvesal 11 months ago - 1 comment

#67 - Use in_chans in patchify and unpatchify

Pull Request - State: open - Opened by ankitkariryaa 11 months ago - 2 comments
Labels: CLA Signed

#66 - deployment issue in trt fp16

Issue - State: open - Opened by sipie800 12 months ago

#65 - Pre-trained weights incompatible with backbone

Issue - State: open - Opened by LucFrachon 12 months ago - 4 comments

#63 - Cannot install MinkowskiEngine with provided instructions

Issue - State: open - Opened by gholste about 1 year ago - 3 comments

#61 - ImageNet-1K pre-trained weights ConvNeXt V2 supervised

Issue - State: open - Opened by OmarEl-Khashab over 1 year ago - 1 comment

#60 - could you share the dense masked conv based sparse encoding ?

Issue - State: open - Opened by hengshan123 over 1 year ago - 2 comments

#58 - add dynamic image size, patch sizes, and image channels

Pull Request - State: open - Opened by lrettenberger over 1 year ago - 3 comments
Labels: CLA Signed

#57 - Make image and patch sizes dynamic

Issue - State: open - Opened by lrettenberger over 1 year ago - 1 comment

#56 - On masking input images

Issue - State: open - Opened by xiaohao-lin1 over 1 year ago - 1 comment

#55 - Code improvement

Pull Request - State: closed - Opened by kytimmylai over 1 year ago - 1 comment
Labels: CLA Signed

#54 - Use Older Version of Apex

Pull Request - State: open - Opened by akhilkedia over 1 year ago
Labels: CLA Signed

#53 - Update INSTALL.md: Typo Fix

Pull Request - State: open - Opened by akhilkedia over 1 year ago - 2 comments
Labels: CLA Signed

#52 - unable to install apex

Issue - State: open - Opened by wgx8130 over 1 year ago - 2 comments

#51 - visualization and feature consine distance

Issue - State: open - Opened by Mogul5306 over 1 year ago - 2 comments

#50 - Unable to update submodule (publickey error)

Issue - State: open - Opened by JonathanBhimaniBurrows over 1 year ago - 4 comments

#49 - segmentation

Issue - State: open - Opened by wangfangming20170245010 over 1 year ago

#48 - The reconstruction images have a lot of noise

Issue - State: closed - Opened by powermano over 1 year ago - 6 comments

#46 - Adding module for pretraining with target as clip embeddings

Pull Request - State: closed - Opened by Charithaakula almost 2 years ago - 1 comment

#45 - Reason for double mask in Jax implementation?

Issue - State: closed - Opened by sebastian-sz almost 2 years ago - 2 comments

#44 - What is the loss of the pretrain model in fcmae stage?

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

#43 - How can I reproduce the supervised training results?

Issue - State: open - Opened by rayleizhu almost 2 years ago - 1 comment

#41 - custom LayerNorm

Issue - State: open - Opened by starsky68 almost 2 years ago

#39 - Could you release a pre-trained model with a decoder?

Issue - State: open - Opened by laibao99 almost 2 years ago - 3 comments

#37 - How about the log files of pretraining?

Issue - State: open - Opened by ayumiymk almost 2 years ago - 1 comment

#35 - cannot import name 'MinkowskiDepthwiseConvolution' from 'MinkowskiEngine'.

Issue - State: closed - Opened by jayepep almost 2 years ago - 5 comments

#34 - How to get the predicted results of the test set?

Issue - State: open - Opened by weihongwei-zg almost 2 years ago

#33 - RuntimeError: Error(s) in loading state_dict for ConvNeXtV2:

Issue - State: open - Opened by weihongwei-zg almost 2 years ago - 5 comments

#31 - Question regarding sparse convolution dimension

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

#30 - About the size of the patches

Issue - State: open - Opened by colindecourt almost 2 years ago - 8 comments

#29 - Reproduction of segmentation networks

Issue - State: open - Opened by 464hee almost 2 years ago

#28 - LayerNorm vs torch.nn.LayerNorm

Issue - State: open - Opened by Michael-H777 almost 2 years ago - 1 comment

#26 - self-supervised pre-trained weights miss Parameters

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

#25 - Question on decoder resolution

Issue - State: closed - Opened by cmsflash almost 2 years ago - 2 comments

#24 - Problem with ConvNeXt V2-B 384x384 weights

Issue - State: closed - Opened by kaitolucifer almost 2 years ago - 5 comments

#23 - Will the methods work on other CNNs?

Issue - State: closed - Opened by dashidhy about 2 years ago - 2 comments

#22 - Is the gamma and beta initial value wrong?Does zero work？

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

#21 - ImageNet-1K Benchmarks (Trained from scratch)

Issue - State: closed - Opened by ahatamiz about 2 years ago - 1 comment

#20 - cd MinkowskiEngine reported an error

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

#19 - parameter shape mismatch in GRN

Issue - State: closed - Opened by MrWaterZhou about 2 years ago - 1 comment

#18 - Update INSTALL.md

Pull Request - State: open - Opened by MJDDH about 2 years ago - 1 comment

#17 - 这个文件夹MinkowskiEngine是空的，没有setup文件

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

#16 - cd MinkowskiEngine报错，没有这个目录文件，是咋回事

Issue - State: closed - Opened by MJDDH about 2 years ago - 1 comment

#15 - Is it difficult to train/finetune ConvNeXtv2 compared with ConvNeXtv1?

Issue - State: open - Opened by linhduongtuan about 2 years ago - 7 comments

#14 - Inplementation mismatch with the paper

Issue - State: closed - Opened by NonBee98 about 2 years ago - 1 comment

#13 - Parameter Settings about COCO Detection

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

#12 - How to use pretrained model in coco detection

Issue - State: closed - Opened by yuangpeng about 2 years ago - 4 comments

#11 - image reconstruction result comparision of the pretrained ConvNeXt-V1/V2 model

Issue - State: open - Opened by songkq about 2 years ago - 4 comments

#10 - Update INSTALL.md

Pull Request - State: closed - Opened by sorrowyn about 2 years ago - 2 comments

#9 - trainning is too slow compared with MAE+VIT-Base

Issue - State: closed - Opened by smarengin about 2 years ago - 3 comments

#8 - can we have the segmentation code?

Issue - State: closed - Opened by keyingjun about 2 years ago - 1 comment

#7 - can we have segmentation code?

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

#6 - Hello, I found some bugs

Issue - State: closed - Opened by wuyx517 about 2 years ago - 1 comment

#5 - Efficiency comparison between dense conv + binary masking and sparse conv

Issue - State: closed - Opened by rayleizhu about 2 years ago - 1 comment

#4 - TF/Keras-based Implementation

Issue - State: open - Opened by DavidLandup0 about 2 years ago - 1 comment

#3 - ConvNeXtV2 TensorFlow implementation

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

#2 - License change from V1

Issue - State: closed - Opened by jturner116 about 2 years ago - 5 comments

#1 - Visualization Question

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