Ina

> @wangrui1573 If you are trying to publish to a server within china/russia [please see this comment](https://github.com/SamKirkland/FTP-Deploy-Action/issues/116#issuecomment-724312343) But why I can deploy my app to the server when it runs...

@ramit-wandb Oh, thank you for your response. Could you please explain to me the second question I asked? > And if I call wandb.finish() manually, how wandb set record_writer_thread.stopped so...

Hi, @ramit-wandb The code snippet that you provided set `stopped` only when `KeyboardInterrupt` has happened, but what confused me was other situations. For example, where wandb set `stopped` when the...

@ramit-wandb Hi, thank you for your patience. But as I see it from https://github.com/wandb/wandb/blob/master/wandb/sdk/backend/backend.py#L246. The user process has to wait for `wandb_process` to finish. I wonder if this causes a...

@ckkelvinchan Hi, I reimplemented the training pipeline by myself, after that, I trained a model with the setting as `mid_channels=64 num_blocks=7 num_input_frames=7` on 4 2080ti. According to my rough estimate,...

> Since the number of frames during training is reduced to 7, I think the time is not very normal. What is the `time` shown in the training log? @ckkelvinchan...

@grimoire Yeah, I found this very helpful, thank you for your timely response! But, I have another question, the last split may have the length of frames smaller than `max_seq_len`,...

@meobach Thanks!

@markohlebar Sorry for disturb you, I found that I hava used wrong path, and I got success when I fix it. Thank you for your quick reply.

@kimlimjustin Yeah, it is a warning, but these are expected results when I run yarn install?