Socrates

EMA Update of bn buffer

5

The following function apply moving average to the ema model. But it didn't update the statistic(runing_mean and runing_var) since these two were not parameters but buffers. ``` def update_moving_average(ema_updater, ma_model,...

The kl_div loss of self distillation

1

The following code calculate the kl_div loss of teacher from stage 1 and the student model. But the student didn't calculate log_softmax. Is this a mistake? ``` if self.opt.distillation >...

## 1.SQM QoS与NAT回环的兼容问题 Q：是否搜索了issue (使用 "x" 选择) * [x] 没有类似的issue ## 2. 详细叙述 ### (1) 具体问题 A：路由器br-lan接口设置了SQM QoS(已正确设置，功能正常)，并且内网有一台nas设置了端口转发和NAT回环，此时内网通过域名访问nas时，如果开启QoS速度会被限制成宽带上行速度，关闭QoS则是正常的内网速度。 (推测是否因为NAT回环是先发送到wan再转发到lan被视作wan口的入站流量，是否有解决方案) ### (2) 路由器型号和固件版本 A：R66S，openwrt-rockchip-R23.1.1-fastrhino_r66s-squashfs-sysupgrade ### (3) 详细日志 A：无

反馈bug/问题模板，提建议请删除 ## 1.关于你要提交的问题 Q：是否搜索了issue (使用 "x" 选择) * [x] 没有类似的issue ## 2. 详细叙述 ### (1) 具体问题 A：SQM 与 sfe 一起开启的情况下，SQM 仅有 **上传速度(egress)** 栏目有效，此时添加2条规则分别限制 lan 和 wan 的情况下可以对内网设备的上传和下载起到保障带宽作用。但是对于端口转发到外网的内网 nas 设备，从外网进行访问时，SQM 无法起到保障带宽的作用。...

I got decent results, but some of the samples were in the wrong color

13

Hello, @guoqincode! The samples in the test set worked fine, but with pure black clothes, I often got black and white results. Have you ever run into similar problems. ...

This repository uses the following code in this [file](https://github.com/rocketappslab/BDMM/blob/main/evaluation/getMetrics_animation.py). But the formula should be `10 * math.log10(255.0 / math.sqrt(mse))`, The result of the calculation is 2 times larger than the...

train_cogvideox_image_to_video_lora bug

2

Thank you for releasing the I2V training code. I noticed an issue in it. At https://github.com/THUDM/CogVideo/blob/main/finetune/train_cogvideox_image_to_video_lora.py#L1279, the information of the image is completely unused. Should it be changed to `noisy_image...

deepspeed时未正确指定bfloat16精度