ziegenbalg

Will report back here once I can confirm gpu training still works. Setting up env for LTS and 1.11.1 this week.

Still broker with gpu I think? (default) [eziegenbalg@localhost-live pytorch-dnc]$ python ./tasks/adding_task.py -cuda 0 -lr 0.0001 -rnn_type lstm -memory_type sam -nlayer 1 -nhlayer 1 -nhid 100 -dropout 0 -mem_slot 1000 -mem_size...

@ixaxaar have you had a chance to see if this works under the new pytorch LTS version?

Hitting this problem as well. ``` Exception ignored in: Traceback (most recent call last): File "/usr/lib64/python3.10/multiprocessing/util.py", line 224, in __call__ res = self._callback(*self._args, **self._kwargs) File "/usr/lib64/python3.10/multiprocessing/synchronize.py", line 87, in _cleanup...

I've tried these changes and get a syntax error: ``` 1 │ export const GA_TRACKING_ID = process.env.NEXT_PUBLIC_GA_ID 2 │ export const pageview = (url: string) => { · ─ 3...

I would like to know this as well. It seems like the keras tuners don't like the tf.data.Datasets yet. They're still expecting `(x_train, y_train), (x_test, y_test).` Is my thinking correct...