zkkli RepQ-ViT issues - Githubissues

zkkli / RepQ-ViT

[ICCV 2023] RepQ-ViT: Scale Reparameterization for Post-Training Quantization of Vision Transformers

Apache License 2.0

102 stars 8 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

reproduced performance is poor in swin-base

#10 XA23i opened 2 months ago
1
Unable to run the code

#9 harshaUwm163 opened 3 months ago
0
What does reconstruction mean in quantization.

#8 XA23i closed 4 months ago
2
I'm sorry, why did I run the example and still get a quantitative model of float32?

#7 HJQjob closed 1 month ago
0
How to understand the concept of channel before layernorm?

#6 xyhe1996 closed 7 months ago
4
Discrepancy between reported and tested mAP for `mask_rcnn_swin_small`

#5 GoatWu opened 8 months ago
0
How to save and load the quantized model?

#4 KaidDuong opened 9 months ago
2
AttributeError: 'Attention' object has no attribute 'reduction'

#3 Arthur-Ling opened 9 months ago
1
Hello! Here's another question about the log sqrt2 quantization.

#2 GoatWu closed 8 months ago
0
Hello! May I inquire if the LayerNorm module has not been quantized in this model?

#1 GoatWu opened 10 months ago
3