Commit 420fb08f authored by Ali Cowen-Rivers's avatar Ali Cowen-Rivers Committed by Copybara-Service
Browse files

Adding recycle information for timings.

PiperOrigin-RevId: 497360470
Change-Id: I3a9f3ba608ac3ceeaaccfe281a82d43fae1d265e
parent a9e5451f
...@@ -353,12 +353,12 @@ can be done via the `--num_multimer_predictions_per_model` flag, e.g. set it to ...@@ -353,12 +353,12 @@ can be done via the `--num_multimer_predictions_per_model` flag, e.g. set it to
### AlphaFold prediction speed ### AlphaFold prediction speed
The table below reports prediction runtimes for proteins of various lengths. We The table below reports prediction runtimes for proteins of various lengths. We
only measure unrelaxed structure prediction while excluding runtimes from MSA only measure unrelaxed structure prediction with three recycles while
and template search. When running `docker/run_docker.py` with excluding runtimes from MSA and template search. When running
`--benchmark=true`, this runtime is stored in `timings.json`. All runtimes are `docker/run_docker.py` with `--benchmark=true`, this runtime is stored in
from a single A100 NVIDIA GPU. Prediction speed on A100 for smaller structures `timings.json`. All runtimes are from a single A100 NVIDIA GPU. Prediction
can be improved by increasing `global_config.subbatch_size` in speed on A100 for smaller structures can be improved by increasing
`alphafold/model/config.py`. `global_config.subbatch_size` in `alphafold/model/config.py`.
No. residues | Prediction time (s) No. residues | Prediction time (s)
-----------: | ------------------: -----------: | ------------------:
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment