GraphOptionsOrBuilder
Sử dụng bộ sưu tập để sắp xếp ngăn nắp các trang
Lưu và phân loại nội dung dựa trên lựa chọn ưu tiên của bạn.
| Các lớp con gián tiếp đã biết |
Phương pháp công cộng
tóm tắt công khai dài getBuildCostModel ()
The number of steps to run before returning a cost model detailing
the memory usage and performance of each node of the graph. 0 means
no cost model.
int64 build_cost_model = 4; tóm tắt công khai dài getBuildCostModelAfter ()
The number of steps to skip before collecting statistics for the
cost model.
int64 build_cost_model_after = 9; boolean trừu tượng công khai getEnableBfloat16Sendrecv ()
If true, transfer float values between processes as bfloat16.
bool enable_bfloat16_sendrecv = 7; boolean trừu tượng công khai getEnableRecvScheduling ()
If true, use control flow to schedule the activation of Recv nodes.
(Currently ignored.)
bool enable_recv_scheduling = 2; boolean trừu tượng công khai getInferShapes ()
Annotate each Node with Op output shape data, to the extent it can
be statically inferred.
bool infer_shapes = 5; Tóm tắt công khai OptimizerOptions getOptimizerOptions ()
Options controlling how graph is optimized.
.tensorflow.OptimizerOptions optimizer_options = 3;
Options controlling how graph is optimized.
.tensorflow.OptimizerOptions optimizer_options = 3; boolean trừu tượng công khai getPlacePrunedGraph ()
Only place the subgraphs that are run, rather than the entire graph.
This is useful for interactive graph building, where one might
produce graphs that cannot be placed during the debugging
process. In particular, it allows the client to continue work in
a session after adding a node to a graph whose placement
constraints are unsatisfiable.
bool place_pruned_graph = 6; tóm tắt công khai RewriterConfig getRewriteOptions ()
Options that control the type and amount of graph rewriting.
Not currently configurable via the public Python API (i.e. there is no API
stability guarantee if you import RewriterConfig explicitly).
.tensorflow.RewriterConfig rewrite_options = 10; tóm tắt công khai RewriterConfigOrBuilder getRewriteOptionsOrBuilder ()
Options that control the type and amount of graph rewriting.
Not currently configurable via the public Python API (i.e. there is no API
stability guarantee if you import RewriterConfig explicitly).
.tensorflow.RewriterConfig rewrite_options = 10; tóm tắt công khai int getTimelineStep ()
If > 0, record a timeline every this many steps.
EXPERIMENTAL: This currently has no effect in MasterSession.
int32 timeline_step = 8; boolean trừu tượng công khai hasOptimizerOptions ()
Options controlling how graph is optimized.
.tensorflow.OptimizerOptions optimizer_options = 3; boolean trừu tượng công khai hasRewriteOptions ()
Options that control the type and amount of graph rewriting.
Not currently configurable via the public Python API (i.e. there is no API
stability guarantee if you import RewriterConfig explicitly).
.tensorflow.RewriterConfig rewrite_options = 10;
Trừ phi có lưu ý khác, nội dung của trang này được cấp phép theo Giấy phép ghi nhận tác giả 4.0 của Creative Commons và các mẫu mã lập trình được cấp phép theo Giấy phép Apache 2.0. Để biết thông tin chi tiết, vui lòng tham khảo Chính sách trang web của Google Developers. Java là nhãn hiệu đã đăng ký của Oracle và/hoặc các đơn vị liên kết với Oracle.
Cập nhật lần gần đây nhất: 2025-07-27 UTC.
[null,null,["Cập nhật lần gần đây nhất: 2025-07-27 UTC."],[],[],null,["# GraphOptionsOrBuilder\n\npublic interface **GraphOptionsOrBuilder** \n\n|---|---|---|\n| Known Indirect Subclasses [GraphOptions](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptions), [GraphOptions.Builder](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptions.Builder) |------------------------------------------------------------------------------------------------|-----------------------------------------| | [GraphOptions](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptions) | Protobuf type `tensorflow.GraphOptions` | | [GraphOptions.Builder](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptions.Builder) | Protobuf type `tensorflow.GraphOptions` | |||\n\n### Public Methods\n\n|-------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| abstract long | [getBuildCostModel](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptionsOrBuilder#getBuildCostModel())() ``` The number of steps to run before returning a cost model detailing the memory usage and performance of each node of the graph. ``` |\n| abstract long | [getBuildCostModelAfter](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptionsOrBuilder#getBuildCostModelAfter())() ``` The number of steps to skip before collecting statistics for the cost model. ``` |\n| abstract boolean | [getEnableBfloat16Sendrecv](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptionsOrBuilder#getEnableBfloat16Sendrecv())() ``` If true, transfer float values between processes as bfloat16. ``` |\n| abstract boolean | [getEnableRecvScheduling](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptionsOrBuilder#getEnableRecvScheduling())() ``` If true, use control flow to schedule the activation of Recv nodes. ``` |\n| abstract boolean | [getInferShapes](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptionsOrBuilder#getInferShapes())() ``` Annotate each Node with Op output shape data, to the extent it can be statically inferred. ``` |\n| abstract [OptimizerOptions](/jvm/api_docs/java/org/tensorflow/proto/framework/OptimizerOptions) | [getOptimizerOptions](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptionsOrBuilder#getOptimizerOptions())() ``` Options controlling how graph is optimized. ``` |\n| abstract [OptimizerOptionsOrBuilder](/jvm/api_docs/java/org/tensorflow/proto/framework/OptimizerOptionsOrBuilder) | [getOptimizerOptionsOrBuilder](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptionsOrBuilder#getOptimizerOptionsOrBuilder())() ``` Options controlling how graph is optimized. ``` |\n| abstract boolean | [getPlacePrunedGraph](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptionsOrBuilder#getPlacePrunedGraph())() ``` Only place the subgraphs that are run, rather than the entire graph. ``` |\n| abstract [RewriterConfig](/jvm/api_docs/java/org/tensorflow/proto/framework/RewriterConfig) | [getRewriteOptions](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptionsOrBuilder#getRewriteOptions())() ``` Options that control the type and amount of graph rewriting. ``` |\n| abstract [RewriterConfigOrBuilder](/jvm/api_docs/java/org/tensorflow/proto/framework/RewriterConfigOrBuilder) | [getRewriteOptionsOrBuilder](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptionsOrBuilder#getRewriteOptionsOrBuilder())() ``` Options that control the type and amount of graph rewriting. ``` |\n| abstract int | [getTimelineStep](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptionsOrBuilder#getTimelineStep())() ``` If \u003e 0, record a timeline every this many steps. ``` |\n| abstract boolean | [hasOptimizerOptions](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptionsOrBuilder#hasOptimizerOptions())() ``` Options controlling how graph is optimized. ``` |\n| abstract boolean | [hasRewriteOptions](/jvm/api_docs/java/org/tensorflow/proto/framework/GraphOptionsOrBuilder#hasRewriteOptions())() ``` Options that control the type and amount of graph rewriting. ``` |\n\nPublic Methods\n--------------\n\n#### public abstract long\n**getBuildCostModel**\n()\n\n\u003cbr /\u003e\n\n```\n The number of steps to run before returning a cost model detailing\n the memory usage and performance of each node of the graph. 0 means\n no cost model.\n \n```\n`int64 build_cost_model = 4;`\n\n\u003cbr /\u003e\n\n#### public abstract long\n**getBuildCostModelAfter**\n()\n\n\u003cbr /\u003e\n\n```\n The number of steps to skip before collecting statistics for the\n cost model.\n \n```\n`int64 build_cost_model_after = 9;`\n\n\u003cbr /\u003e\n\n#### public abstract boolean\n**getEnableBfloat16Sendrecv**\n()\n\n\u003cbr /\u003e\n\n```\n If true, transfer float values between processes as bfloat16.\n \n```\n`bool enable_bfloat16_sendrecv = 7;`\n\n\u003cbr /\u003e\n\n#### public abstract boolean\n**getEnableRecvScheduling**\n()\n\n\u003cbr /\u003e\n\n```\n If true, use control flow to schedule the activation of Recv nodes.\n (Currently ignored.)\n \n```\n`bool enable_recv_scheduling = 2;`\n\n\u003cbr /\u003e\n\n#### public abstract boolean\n**getInferShapes**\n()\n\n\u003cbr /\u003e\n\n```\n Annotate each Node with Op output shape data, to the extent it can\n be statically inferred.\n \n```\n`bool infer_shapes = 5;`\n\n\u003cbr /\u003e\n\n#### public abstract [OptimizerOptions](/jvm/api_docs/java/org/tensorflow/proto/framework/OptimizerOptions)\n**getOptimizerOptions**\n()\n\n\u003cbr /\u003e\n\n```\n Options controlling how graph is optimized.\n \n```\n`.tensorflow.OptimizerOptions optimizer_options = 3;`\n\n\u003cbr /\u003e\n\n#### public abstract [OptimizerOptionsOrBuilder](/jvm/api_docs/java/org/tensorflow/proto/framework/OptimizerOptionsOrBuilder)\n**getOptimizerOptionsOrBuilder**\n()\n\n\u003cbr /\u003e\n\n```\n Options controlling how graph is optimized.\n \n```\n`.tensorflow.OptimizerOptions optimizer_options = 3;`\n\n\u003cbr /\u003e\n\n#### public abstract boolean\n**getPlacePrunedGraph**\n()\n\n\u003cbr /\u003e\n\n```\n Only place the subgraphs that are run, rather than the entire graph.\n This is useful for interactive graph building, where one might\n produce graphs that cannot be placed during the debugging\n process. In particular, it allows the client to continue work in\n a session after adding a node to a graph whose placement\n constraints are unsatisfiable.\n \n```\n`bool place_pruned_graph = 6;`\n\n\u003cbr /\u003e\n\n#### public abstract [RewriterConfig](/jvm/api_docs/java/org/tensorflow/proto/framework/RewriterConfig)\n**getRewriteOptions**\n()\n\n\u003cbr /\u003e\n\n```\n Options that control the type and amount of graph rewriting.\n Not currently configurable via the public Python API (i.e. there is no API\n stability guarantee if you import RewriterConfig explicitly).\n \n```\n`.tensorflow.RewriterConfig rewrite_options = 10;`\n\n\u003cbr /\u003e\n\n#### public abstract [RewriterConfigOrBuilder](/jvm/api_docs/java/org/tensorflow/proto/framework/RewriterConfigOrBuilder)\n**getRewriteOptionsOrBuilder**\n()\n\n\u003cbr /\u003e\n\n```\n Options that control the type and amount of graph rewriting.\n Not currently configurable via the public Python API (i.e. there is no API\n stability guarantee if you import RewriterConfig explicitly).\n \n```\n`.tensorflow.RewriterConfig rewrite_options = 10;`\n\n\u003cbr /\u003e\n\n#### public abstract int\n**getTimelineStep**\n()\n\n\u003cbr /\u003e\n\n```\n If \u003e 0, record a timeline every this many steps.\n EXPERIMENTAL: This currently has no effect in MasterSession.\n \n```\n`int32 timeline_step = 8;`\n\n\u003cbr /\u003e\n\n#### public abstract boolean\n**hasOptimizerOptions**\n()\n\n\u003cbr /\u003e\n\n```\n Options controlling how graph is optimized.\n \n```\n`.tensorflow.OptimizerOptions optimizer_options = 3;`\n\n\u003cbr /\u003e\n\n#### public abstract boolean\n**hasRewriteOptions**\n()\n\n\u003cbr /\u003e\n\n```\n Options that control the type and amount of graph rewriting.\n Not currently configurable via the public Python API (i.e. there is no API\n stability guarantee if you import RewriterConfig explicitly).\n \n```\n`.tensorflow.RewriterConfig rewrite_options = 10;`\n\n\u003cbr /\u003e"]]
