Search Options

Results per page
Sort
Preferred Languages
Advance

Results 1 - 4 of 4 for pessimistic (0.42 sec)

  1. tensorflow/compiler/jit/deadness_analysis.cc

          // The optimistic mode does not converge.  Let's fall back to the
          // pessimistic mode.
          TF_RETURN_IF_ERROR(
              PopulateFrame(sub_topo, /*use_optimistic_mode=*/false, nullptr));
        }
        VLOG(2) << "Done populating frame " << cur_frame_name << " using the "
                << (success ? "optimistic" : "pessimistic") << " mode.";
      }
    
      return absl::OkStatus();
    }
    
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Tue Mar 12 06:33:33 UTC 2024
    - 60.4K bytes
    - Viewed (0)
  2. tensorflow/compiler/mlir/tensorflow/analysis/resource_alias_analysis.cc

        // calls a function that has not yet been analyzed, we will not be able to
        // backtrack through that function call (our analysis will be correct but
        // pessimistic).
        for (CallGraphNode* node : scc) {
          if (node->isExternal()) continue;
          Region* region = node->getCallableRegion();
          GetOrCreateAnalysis(*region);
        }
      }
    
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Wed May 15 09:04:13 UTC 2024
    - 28.2K bytes
    - Viewed (0)
  3. tensorflow/compiler/mlir/tf2xla/internal/passes/extract_outside_compilation.cc

          // cluster that will be moved to the host at the same time since both
          // defining op and user op will be moved to host.
          if (cluster_ops.count(user)) {
            continue;
          }
          // This is pessimistic and in some cases will add extra communication.
          if (!HasOutsideCompilationAncestor(user) || has_dynamic_outputs ||
              HasDynamicOutputs(user)) {
            if (!user_set.insert(user).second) continue;
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Tue Apr 30 21:25:12 UTC 2024
    - 68.3K bytes
    - Viewed (0)
  4. src/runtime/mgcscavenge.go

    		// that the overheads of scavenging are better amortized, so better
    		// scavenging throughput.
    		//
    		// The current value is chosen assuming a cost of ~10µs/physical page
    		// (this is somewhat pessimistic), which implies a worst-case latency of
    		// about 160µs for 4 KiB physical pages. The current value is biased
    		// toward latency over throughput.
    		const scavengeQuantum = 64 << 10
    
    Registered: Wed Jun 12 16:32:35 UTC 2024
    - Last Modified: Wed May 08 17:48:45 UTC 2024
    - 52.3K bytes
    - Viewed (0)
Back to top