Issues
- 1
Question about cls token
#100 opened by eremo2002 - 4
- 0
How to switch to the VOC2007 dataset
#116 opened by Esther-Mo - 1
CityScapes和CamVid数据集训练
#115 opened by miscedence12 - 0
About loading info when load pvt_v2_b2.pth with training config retinanet_pvt_v2_b2_fpn_1x_coco.py
#114 opened by YangGangZhiQi - 1
为什么没有办法复现论文中的精度?
#111 opened by shy-star - 1
Visualizing Attention Map
#80 opened by chandlerbing65nm - 0
figure 2 in paper (PVT)
#113 opened by fyting - 0
Cannot download detr_pvt_s_8x2_50ep_coco.pth checkpoint model from google drive link
#112 opened by Hule95 - 4
About the segmentation and detection. I use the imagenet-1k pretrained model to segmentation and detection task. but the mIoU I get is very low
#88 opened by 1104662797 - 0
Porting model into hugging face transformers.
#110 opened by Xrenya - 0
Required to update dataset preparation link
#109 opened by StuffyMonkey - 0
Welcome update to OpenMMLab 2.0
#108 opened by vansin - 0
In the PVT-v2 paper, shouldn't the time complexity for SRA be [2*(h^2)*(w^2)*c]/(R^2) + hw(c^2) instead?
#107 opened by Phuoc-Hoan-Le - 0
In the PVT-v2 code, have you tried not using a linear projection after the pooling layer in the spatial reduction part of the attention?
#106 opened by Phuoc-Hoan-Le - 0
KeyError: 'MaskRCNN: "pvt_v2_b1: \'pretrained\'"'
#105 opened by yuhua666 - 0
Why there is no DETR+PVTv2 in object detection?
#104 opened by yuhua666 - 0
- 0
Did you train PVT on ImageNet22k?
#102 opened by Roger-Liang - 1
patch_embed related questions.
#85 opened by zijunwei - 2
Vision Language examples
#101 opened by AwePhD - 0
Pre-training code
#99 opened by nargenziano - 0
without Convolutions?
#98 opened by Oguzhanercan - 0
Question about pooling size
#97 opened by magehrig - 0
In detection, how to test the test datasets?
#96 opened by jieruyao49 - 0
log for pvt v2 classification case
#95 opened by zhicheng2T0 - 2
pvt-v2 for segmentation tasks
#86 opened by zijunwei - 0
Gradient overflow. Skipping step, loss scaler 0 reducing loss scale to 0.03125
#94 opened by Williamlizl - 2
PVT for Faster RCNN
#81 opened by chandlerbing65nm - 0
image size 384
#93 opened by xmy0916 - 0
learning rate for AdamW
#92 opened by chandlerbing65nm - 2
The training time of PVT
#83 opened by 1104662797 - 0
Fine tuning!
#91 opened by ireneb612 - 1
Questions on PVT + FPN
#87 opened by DianCh - 3
question for PVTv2: in the paper the reduction ratio is 7 in Linear SRA, but in the code is sr_ratios=[8, 4, 2, 1]
#90 opened by StormArcher - 0
Why the lr is too small?
#89 opened by DotWang - 1
- 1
I have trained your pvt_tiny model for 300 epochs, but I got 152m checkpoint.pth, but your model just 51m. Is there anything wrong?
#75 opened by 1104662797 - 0
about the parameters of the model
#79 opened by kingkung2016 - 0
PVTv1 classification train log
#78 opened by ManLoveRice - 2
About `patch_size` in pvt_v2
#72 opened by abhi-glitchhg - 1
- 0
SRA part of the paper description is inconsistent with the code implementation
#74 opened by torchmyheart - 0
The prolem of get_flops.py
#73 opened by ucasligang - 0
RuntimeError: The size of tensor a (1000) must match the size of tensor b (512) at non-singleton dimension 1
#69 opened by ersamo - 0
some question about init_weights
#70 opened by Yangr116 - 1
是否能够提供classification训练过程中的acc,或者训练过程中的损失曲线图?
#66 opened by 1104662797 - 0
- 5
Use PVTv2 as backbone for custom model
#65 opened by EckoTan0804 - 0