4 commits

Author	SHA1	Message	Date
Richard Liaw	7e62e1187c	[air/benchmark] Torch benchmarks for 4x4 (#26692) ... Add benchmark data for 4x4 GPU setup. Signed-off-by: Richard Liaw <rliaw@berkeley.edu> Co-authored-by: Jimmy Yao <jiahaoyao.math@gmail.com> Co-authored-by: Kai Fricke <kai@anyscale.com>	2022-07-19 17:06:37 +01:00
Kai Fricke	00947fd949	[air/benchmarks] Add 4x1 GPU benchmark for Torch (#26562)	2022-07-18 12:14:10 -07:00
Kai Fricke	213a96e239	[air/benchmarks] Add distributed Tensorflow benchmarks (CPU only) (#26519) ... Following up from #26436, this PR adds a distributed benchmark test for Tensorflow FashionMNIST training. It compares training with Ray AIR with training with vanilla PyTorch. Signed-off-by: Kai Fricke <kai@anyscale.com>	2022-07-14 22:08:43 +01:00
Kai Fricke	cf75cf7232	[air] Add AIR distributed training benchmark for Torch FashionMNIST (#26436) ... This PR adds a distributed benchmark test for Pytorch MNIST training. It compares training with Ray AIR with training with vanilla PyTorch. In both cases, the same training loop is used. For Ray AIR, we use a TorchTrainer with 4 CPU workers. For vanilla PyTorch, we upload a training script and kick it off (using Ray tasks) in subprocesses on each node. In both cases, we collect the end to end runtime. Signed-off-by: Kai Fricke <kai@anyscale.com>	2022-07-13 10:53:24 +01:00