Training Run Report

1
Rounds
2
Workers
9.9s
Total Time
1.6MB
Data Transferred
799.4KB
Params Pushed
799.4KB
Weights Pulled
8.7
Final |W|
Push params
Train
Pull weights
Aggregate
Dead / Error
round 0 [2/2] 9.91s 0.0s 2.0s 4.0s 5.9s 7.9s 9.9s scheduler push 399.7KB → node 0 (0.145s) push 399.7KB → node 1 (0.146s) pull 399.7KB ← node 0 (0.174s) pull 399.7KB ← node 1 (0.165s) aggregate: 2 workers, |W|=8.72, Δ=0.3177 worker 0 push 399.7KB → node 0 (0.145s) train node 0: 9.42s exit=0 cpu = _conversion_method_template(device=torch.device("cpu")) pull 399.7KB ← node 0 (0.174s) worker 1 push 399.7KB → node 1 (0.146s) train node 1: 9.42s exit=0 cpu = _conversion_method_template(device=torch.device("cpu")) pull 399.7KB ← node 1 (0.165s)
[   0.15s] exec  node=1  exit=0    0.142s  | Warning: Permanently added 'worker-1' (ED25519) to the list of known hosts.
[   0.15s] exec  node=0  exit=0    0.144s  | Warning: Permanently added 'worker-0' (ED25519) to the list of known hosts.
[   0.30s] push  node=1  job_module.py           1.4KB  0.146s
[   0.30s] push  node=0  job_module.py           1.4KB  0.146s
[   0.30s] ──── round 0 start (2 workers) ────────────────
[   0.45s] push  node=0  params.pt             399.7KB  0.145s
[   0.45s] push  node=1  params.pt             399.7KB  0.146s
[   9.86s] exec  node=1  exit=0    9.416s  | cpu = _conversion_method_template(device=torch.device("cpu"))
[   9.87s] exec  node=0  exit=0    9.419s  | cpu = _conversion_method_template(device=torch.device("cpu"))
[  10.04s] pull  node=0  weights.pt            399.7KB  0.174s
[  10.21s] pull  node=1  weights.pt            399.7KB  0.165s
[  10.21s] agg   2 workers  |W|=8.72  Δ=0.3177  0.002s
[  10.21s] ──── round 0 end (2/2 survived, 9.91s) ────────