Search Options

Results per page
Sort
Preferred Languages
Advance

Results 1 - 1 of 1 for lgamma (0.59 sec)

  1. tensorflow/compiler/mlir/lite/stablehlo/transforms/composite_avg_pool.cc

        return row * pool.w_out + col;
      };
    
      std::vector<float> correction_data(out_shaped_type.getNumElements(), 1.0);
    
      const float kern_size = pool.kh * pool.kw;
    
      // LEMMA 1: Changing the rounding mode from floor to ceil will increase an
      // output dimension by at most 1 (see `ComputeSpatialOutDim`). This is because
      // for any `x`, `ceil(x) - floor(x) <= 1`.
    
    Registered: Sun Jun 16 05:45:23 UTC 2024
    - Last Modified: Tue May 28 23:16:05 UTC 2024
    - 9.2K bytes
    - Viewed (0)
Back to top