): 2.0.1 (True) - PyTorch XPU available: False - PyTorch NPU available: False - System RAM: 125.74 GB - GPU type: NVIDIA GeForce RTX 3090 - `Accelerate` default config: - compute_environment: LOCAL_MACHINE - distributed_type: FSDP - mixed_precision: fp16 - use_cpu: False - debug:...
(default: None) --epoch-interval EPOCH_INTERVAL Interval of epochs between calculating most convergence heuristics (default: 5) UMAP calculation arguments: --force-umap-cpu Override default UMAP GPU-bound implementation via cuML to use umap-learn library instead (default: False) --subset SUBSET Ma...