COMPUTING APPARATUS

• Main Authors: OSHIMA TAKASHI, TAKAHASHI MAKOTO
• Format: Patent
• Language: English; Japanese
• Published: 27.03.2024
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

To provide a computing apparatus capable of appropriately performing DNN (Deep Neural Network) computation.SOLUTION: A computing apparatus includes at least one computing device including a computation area. The computing apparatus acquires input information, generates a plurality of input features from the input information, and performs DNN computation in parallel on the generated plurality of input feature maps by at least one DNN partitioning method including channel partitioning. In the channel partitioning, by grouping the feature maps into sets of the feature maps without partitioning the feature maps and allocating each set of the feature maps to each computation area, computation for layers is performed.SELECTED DRAWING: Figure 3 【課題】好適にＤＮＮ(Deep Neural Network：深層学習)の演算を行うことができる、演算装置を提供すること。【解決手段】演算領域を有する演算デバイスを少なくとも１つ備え、入力情報を取得すると、入力情報から複数の入力の特徴マップを生成し、生成した複数の入力の特徴に対して、チャンネル分割を含む少なくとも１つのＤＮＮの分割方式で、ＤＮＮの演算を並列で行い、チャンネル分割は、特徴マップを分割することなく、特徴マップをまとめた組に分けて、分けた特徴マップをまとめた組それぞれを、演算領域それぞれに割り当てて、層の演算を行う。【選択図】図３