Fabio De Simone

this issue can also occur while using encode.py, in load_dataset.py, and can be solved in the same way by adding `encoding="utf-8"` in the `load_dataset()` function ` # Plain text with...

i'm having the same problem trying to train the 355M model on a RTX2070 8GB, even with both `--memory_saving_gradients` and `--optimizer sgd` i get the following error ``` tensorflow.python.framework.errors_impl.ResourceExhaustedError: OOM...

this seems to still happen on the latest commit even though the pr was merged my error seems slightly different from the one in this post, but the exception is...

dheavy's pr got merged, but the excepttion is still there albeit with some new errors, posting the new stacktraces ``` > poetry run 01 --local ○ Starting... ▌ 01 is...

Ok so apparently piper got fixed, but the exception above, the one about the source.clients.auto module still exists, as far as i understand this is a separate issue then? should...

```2023-07-24 11:57:50.390 +02:00 [FTL] [PROGRAM] FirstChanceException: Unable to read data from the transport connection: The I/O operation has been aborted because of either a thread exit or an application request.....

I've been trying to artificially recreate this to no avail, it must be something very specific with my setup, what's been happening is that we are having very spotty power...

Ok it's happened again just now, i have the program open and mqtt is actually stopped  logs seem more of the same stuff i already had (extended logging is...

Yes i am using mdns for the server URI, i will try setting the ip instead and try it for a couple days to see if it stops happening just...

> I have strange question, are you by any chance using mDNS? (We're using one HttpClient per whole app so I'm thinking about DNS resolution) Update: i've been since using...