- Sort Score
- Result 10 results
- Languages All
Results 1 - 6 of 6 for graph_op (0.12 sec)
-
tensorflow/compiler/jit/encapsulate_subgraphs_pass.cc
return results_.at(OutputTensor(edge->src(), edge->src_output())); } Node* Encapsulator::Subgraph::MakeNodeImage(const Graph* graph_in, Node* node) { if (!graph_) { graph_.reset(new Graph(graph_in->op_registry())); graph_->set_versions(graph_in->versions()); } // TODO(b/116981129): Enhance how the device for the encapsulated subgraph is
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu Feb 22 08:47:20 UTC 2024 - 51K bytes - Viewed (0) -
tensorflow/compiler/mlir/tensorflow/tests/convert_control_to_data_outputs.mlir
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Thu Feb 22 18:35:00 UTC 2024 - 68.9K bytes - Viewed (0) -
tensorflow/compiler/jit/mark_for_compilation_pass.cc
Cluster* GetClusterForCyclesGraphNode(int node_id) { // We have to check `graph_->FindNodeId(node) == nullptr` because we add all // nodes in [0, graph_->num_node_ids()) to the cycle detection graph but the // TF graph may be missing some node ids. if (node_id >= graph_->num_node_ids() || graph_->FindNodeId(node_id) == nullptr) { return nullptr; }
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Feb 21 12:19:41 UTC 2024 - 85.3K bytes - Viewed (0) -
tensorflow/c/c_api_test.cc
protected: CApiColocationTest() : s_(TF_NewStatus()), graph_(TF_NewGraph()) {} void SetUp() override { feed1_ = Placeholder(graph_, s_, "feed1"); ASSERT_EQ(TF_OK, TF_GetCode(s_)) << TF_Message(s_); feed2_ = Placeholder(graph_, s_, "feed2"); ASSERT_EQ(TF_OK, TF_GetCode(s_)) << TF_Message(s_); constant_ = ScalarConst(10, graph_, s_); ASSERT_EQ(TF_OK, TF_GetCode(s_)) << TF_Message(s_);
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Mon Apr 15 03:35:10 UTC 2024 - 96.9K bytes - Viewed (0) -
tensorflow/compiler/jit/deadness_analysis.cc
absl::flat_hash_map<absl::string_view, size_t> num_enters_for_frame; absl::flat_hash_map<absl::string_view, size_t> num_exits_for_frame; std::vector<size_t> num_ready_inputs(graph_.num_node_ids(), 0); Node* src_node = graph_.source_node(); for (const auto* node : graph_.op_nodes()) { const ControlFlowInfo& cf = control_flow_info_[node->id()]; if (IsRootEnter(node)) {
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Tue Mar 12 06:33:33 UTC 2024 - 60.4K bytes - Viewed (0) -
tensorflow/compiler/mlir/g3doc/_includes/tf_passes.md
-ops-to-preserve : Comma separated list of ops that should not be pruned regardless of reachability ``` ### `-tf-executor-island-coarsening` _Walks tf_executor::GraphOp and merges individual tf_executor::IslandOps._ This pass performs whole graph analysis for a graph encapsulated into tf_executor::GraphOp. The analysis identifies all IslandOps within the graph which could be merged together. The goal is to merge as many islands as possible.
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed Aug 02 02:26:39 UTC 2023 - 96.4K bytes - Viewed (0)