Repository navigation
Port mgpu gqe - #201
Port mgpu gqe#201
Conversation
Signed-off-by: Melody Ren <melodyr@nvidia.com>
Signed-off-by: Melody Ren <melodyr@nvidia.com>
Signed-off-by: Melody Ren <melodyr@nvidia.com>
Signed-off-by: Melody Ren <melodyr@nvidia.com>
Signed-off-by: Melody Ren <melodyr@nvidia.com>
Signed-off-by: Melody Ren <melodyr@nvidia.com>
|
/ok to test 6c4f212 |
Signed-off-by: Melody Ren <melodyr@nvidia.com>
Signed-off-by: Melody Ren <melodyr@nvidia.com>
|
/ok to test 0142ed6 |
|
/ok to test 55275a7 |
Signed-off-by: Melody Ren <melodyr@nvidia.com>
|
/ok to test 577b6af |
|
Where is the log file of the gqe is saved? User might need to check the training performance. Currently, loss value is not provided. Therefore, it is hard to know if the model converged or not. |
|
I tried this example below which represents the H2 molecule. Each time I run it, it gives different results. This was not happening with old version. Output was -1.137. |
I'll look into this and also add logging for the loss value. |
|
Hi @melody-ren |
Signed-off-by: Melody Ren <melodyr@nvidia.com>
|
/ok to test 834e286 |
Signed-off-by: Melody Ren <melodyr@nvidia.com>
|
/ok to test a431984 |
|
/ok to test ca0874d |
Signed-off-by: Melody Ren <melodyr@nvidia.com>
|
/ok to test a8681eb |
Signed-off-by: Melody Ren <melodyr@nvidia.com>
|
/ok to test 0917714 |
Signed-off-by: Melody Ren <melodyr@nvidia.com>
|
/ok to test 8211b74 |
Signed-off-by: Melody Ren <melodyr@nvidia.com>
|
/ok to test 7e3ad67 |
|
Could you also add notes may be as comment in the h2 example on what user should do if they just want to run it on single node multi-GPU. Currently, only for multi-node multi-GPU using mpi. |
|
/ok to test a489590 |
Sure. Will address this in a follow-up PR. Tracked by issue #229 |
Signed-off-by: Melody Ren <melodyr@nvidia.com>
No description provided.