tensorflow/core/profiler/g3doc/options.md
For all tfprof views, the profiles are processed with the following procedures
An in-memory data structure is built represent the view.
-account_type_regexes is used to first select the nodes that includes
the specified operation types. An operation has its default type
(e.g. MatMul, Conv2D). tfprof also considers device as operation type.
User can also define customized operation type. Hence, an operation has
multiple types. Profiler nodes containing matched
types are selected for display and their statistics are aggregated by the
parents of the in-memory data structure.
Various -xxx_name_regexes, -min_xxx, -max_depth etc options are then
applied to further filter based on profiler node names and statistics.
It's no limited operation name. In code view,
it's the code string. In op view, it's the operation type name. Different
from -account_type_regexes, Statistics are used even if a profiler node is not displayed.
For example, in code view, a callee might be hidden, but its statistics is
still aggregated by it's caller. -account_displayed_op_only, however,
breaks the rule and only aggregates statistics of displayed names.
Finally, the filtered data structure is output in a format depending
on the -output option.
options usually have the same semantics in different views. However, some
can vary. For example -max_depth in scope view means the depth of
name scope <b>tree</b>. In op view, it means the length of operation <b>list</b>.
In graph view, in means the number of hops in the <b>graph</b>.
Most machines have multi-core CPUs. Some installs one or more accelerators. Each accelerator usually performs massive parallel processing. The profiler tracks the accumulated processing times. Hence, the accumulated processing time is likely larger than the time of each step.
micros: This is the sum of cpu and accelerator times. accelerator_micros: This is the accelerator times. cpu_micros: This is the cpu times.
Tensor memory are usually ref-counted. The memory is released when there is no more reference to it. It will be difficult to track the release of memory. Currently, profiler only tracks the allocation of memory. As a result, the accumulated memory request is usually larger than the peak memory of the overall model.
It's recommended to generate timeline to see the allocator memory usage over time.
bytes: The memory allocations requested by the operation.
peak_bytes: The peak requested memory (not de-allocated) by the operation.
residual_bytes: The memory requested by the operation and not de-allocated
when Compute finishes.
output_bytes: The memory output by the operation. It's not necessarily requested
by the current operation. For example, it can be a tensor
forwarded from input to output, with in-place mutation.
-max_depth: Show nodes that are at most this number of hops from starting node in the data structure.
-min_bytes: Show nodes that request at least this number of bytes.
-min_peak_bytes: Show nodes that using at least this number of bytes during peak memory usage.
-min_residual_bytes: Show nodes that have at least this number of bytes not being de-allocated after Compute.
-min_output_bytes: Show nodes that have at least this number of bytes output (no necessarily allocated by the nodes).
-min_micros: Show nodes that spend at least this number of microseconds to run. It sums
accelerator_micros and cpu_micros. Note: cpu and accelerator can run in parallel.
-min_accelerator_micros: Show nodes that spend at least this number of microseconds to run on accelerator (e.g. GPU).
-min_cpu_micros: Show nodes that spend at least this number of microseconds to run on CPU.
-min_params: Show nodes that contains at least this number of parameters.
-min_float_ops: Show nodes that contain at least this number of float operations. Only available if an node has op.RegisterStatistics() defined and OpLogProto is provided
-min_occurrence: Show nodes that appear at least this number of times..
-step: Show the stats of the this step when multiple steps of RunMetadata were added. By default, show the average of all steps."
-order_by: Order the results by [name|depth|bytes|peak_bytes|residual_bytes|output_bytes|micros|accelerator_micros|cpu_micros|params|float_ops|occurrence]
-account_type_regexes: Account and display the nodes whose types match one of
the type regexes specified. tfprof allow user to define extra operation types
for graph nodes through tensorflow.tfprof.OpLogProto proto. regexes are
comma-separated.
-start_name_regexes: Show node starting from the node that matches the regexes, recursively. regexes are comma-separated.
-trim_name_regexes: Hide node starting from the node that matches the regexes,
recursively, regexes are comma-separated.
-show_name_regexes: Show node that match the regexes. regexes are
comma-separated.
-hide_name_regexes: Hide node that match the regexes. regexes are
comma-separated.
-account_displayed_op_only: If True, only account the statistics of ops eventually displayed. If False, account all op statistics matching -account_type_regexes recursively.
Notes: See <b>overview</b> session on how does above options play with each other to decide the output and counting.
-select: Comma-separated list of attributes to show. Supported attributes:
[bytes|peak_bytes|residual_bytes|output_bytes|micros|accelerator_micros|cpu_micros|params|float_ops|occurrence|tensor_value|device|op_types|input_shapes].
-output: Output results as stdout, file or timeline.
The format is output_type:key=value,key=value.
For example: -output timeline:outfile=<filename>.
timeline: key=outfile, value=<filename>.
stdout: none.
file: key=outfile, value=<filename>.