-
It will be really nice if you could provide us the code for MAE pretraining.
-
Hello, I have a question about the method of calculating mae. The below is coded in your code.
"""
+= torch.sum(torch.min(torch.stack((torch.abs(p_gt_deg - p_pred_deg), torch.abs(p_pred_deg + 360 …
-
Hello dear authors, thanks for your amazing work!
I find the current code use Point-M2AE as backbone, can you release the code use Point-MAE as backbone?
Thanks!!!
-
Hi author, I would like to ask when calculating depth_mae, why is the predicted depth value less than or equal to zero set to 40. what is the purpose of doing this?
’evaluation_scripts/evaluate_sca…
-
I replaced the Linear Layer in the prediction part with a traditional transformer, but the results were not satisfactory (ETTh1 dataset). The performance is far from the indicators of the transformer+…
-
![屏幕截图 2024-04-27 024826](https://github.com/lijunxian111/UIGNN/assets/65492300/9eee54b9-c212-4ba1-9f35-864c4bf66310)
-
I have recently set up Whisper Writer, and it appears very promising. However, I am facing an issue with inconsistent transcription quality.
In some recordings, Whisper Writer transcribes everythin…
-
你好,结合论文和代码顺利把代码run起来了,对于median有一些疑问,median是在collect和test阶段产生,分别用于指导fine-tuning基于50%分位数进行迭代和test infer的point forecast。代码中生成median的部分如下:
true = np.array(true)[96-pred_len:]
gen_samples = np.array(pred…
-
May I ask the author, why is the RMSE written in the evaluation index in the paper, but the MAE is calculated on the evaluation code?
-
test.py run result:
model load successfully
Adaptive adjacency matrix generation is skipped as 'nodevec1' or 'nodevec2' is not initialized.
Evaluate best model on test data for horizon 1, Test MAE:…