Man1978-scd

train speed is very lower

2

When I use lightGBM on the server, the training speed is very slow, the CPU and memory are sufficient, the data volume is about 2w, even slower than my own...

How to calculate yhat_upper and yhat_lower like prophet, divide the data set into training set and validation set, and then use the validation error to calculate the confidence interval of...

LightGBM Rank Model predict score is negative

2

When I do a rank task, I use the lightgbm ranker model, the loss function uses lambdarank, the data labels are 0 and 1, representing the correlation, when I do...

运行报错

1

当我使用torch1.10.0的时候 执行训练脚本 `bash tools/dist_train.sh work_configs/tamper/tamper_convx_b_exp.py 2` 报错如下： `RuntimeError: one of the variables needed for gradient computation has been modified by an inplace operation: [torch.cuda.HalfTensor [5, 512, 32, 32]], which is...