ag_forward_profile: Profile the forward pass by stage
In ggmlR: 'GGML' Tensor Operations for Machine Learning
View source: R/ag_fwd_profile.R
ag_forward_profile R Documentation
Profile the forward pass by stage
Description
Records where each ag_* operation spends its time on the GPU path:
context handling, tensor creation, buffer allocation, operand upload, graph
building, compute, and result download.
Usage
ag_forward_profile(on = TRUE)
ag_forward_profile_reset()
ag_forward_profile_report()
Arguments
on
'TRUE' to record, 'FALSE' to stop, 'NA' to query.
Details
The upload and download stages are what a resident forward would remove, so
their share is the ceiling for that work – measured rather than estimated.
Value
The previous state, invisibly.
ggmlR documentation built on Oct. 23, 2026, 1:07 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
View source: R/ag_fwd_profile.R
| ag_forward_profile | R Documentation |
Profile the forward pass by stage
Description
Records where each ag_* operation spends its time on the GPU path:
context handling, tensor creation, buffer allocation, operand upload, graph
building, compute, and result download.
Usage
ag_forward_profile(on = TRUE)
ag_forward_profile_reset()
ag_forward_profile_report()
Arguments
on |
'TRUE' to record, 'FALSE' to stop, 'NA' to query. |
Details
The upload and download stages are what a resident forward would remove, so their share is the ceiling for that work – measured rather than estimated.
Value
The previous state, invisibly.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.