James Betker 93a3302819 Push training_state data to CPU memory before saving it ... For whatever reason, keeping this on GPU memory just doesn't work. When you load it, it consumes a large amount of GPU memory and that utilization doesn't go away. Saving to CPU should fix this.		2022-03-04 17:57:33 -07:00
..
__init__.py
audio_resampler.py	Stop dataset - attempt #2	2021-08-18 18:29:38 -06:00
audio.py	misc nonfunctional	2021-11-22 17:16:39 -07:00
colors.py	Add FDPL Loss	2020-07-30 20:47:57 -06:00
convert_model.py	Add "dataset_debugger" support	2022-01-06 12:38:20 -07:00
distributed_checkpont.py	Add distributed_checkpoint for more efficient checkpoints	2020-10-06 20:38:38 -06:00
gpu_mem_track.py	Add GPU mem tracing module	2020-06-14 11:02:54 -06:00
kmeans.py	test uresnet playground mods	2021-01-23 13:46:43 -07:00
loss_accumulator.py	Move log consensus to train for efficiency	2022-03-04 13:41:32 -07:00
numeric_stability.py	More refactor changes	2020-12-18 09:24:31 -07:00
options.py	Fix options.py bug	2021-10-29 14:47:31 -06:00
util.py	Push training_state data to CPU memory before saving it	2022-03-04 17:57:33 -07:00
weight_scheduler.py	Add scheduling to quantizer, enable cudnn_benchmarking to be disabled	2021-09-24 17:01:36 -06:00