Liu Jin
Liu Jin
> How does the code in the psconv.py file achieve the differentdilation rates written in the paper? I think the different dilation rates are done through the three different convs....
Same request here, possible to release training code? It would be extremely helpful.
Any body wonders why the original implementation in MXNet just used a mx.sym.cos(), not giving consideration to (theta + m) > pi?
> How you obtained emore dataset? Did you perform all steps from "Data preparation" instruction in README.md file? In terms of the last step, can you give an example command?...
@siriuz42 Just a quick check, should I set return_forecast_on_context=True for finetuning? I presume it has been set to True for pretraining, but would it be necessary for finetuning?
> @doitslow if you have figured out how to do finetuning on pytorch, can you please share your code? I can't share code. But here is what I did: -...
> @doitslow Thank you for your help. Can you please explain what are they trying to do here in below code? [timesfm_link](https://github.com/google-research/timesfm/blob/ff2a350ba317f462f62485b523d16198a2508c22/src/timesfm/timesfm_torch.py#L122C5-L135C1) Also, do I need to do what they...
@ani0135 You would need mean_output for loss calculation.