AMD-SHARK-Studio/shark/examples/shark_inference/minilm_tf_gpu_config.json

{"options": [{"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 64, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 32, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 64, 32], "work_group_sizes": [128, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 64, 64, 32], "work_group_sizes": [128, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4, "split_k": 8}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 64, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 32, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 64, 32], "work_group_sizes": [128, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 64, 64, 32], "work_group_sizes": [128, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4, "split_k": 8}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 64, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 32, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 64, 32], "work_group_sizes": [128, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 64, 64, 32], "work_group_sizes": [128, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4, "split_k": 8}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 64, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 32, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 64, 32], "work_group_sizes": [128, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 64, 64, 32], "work_group_sizes": [128, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4, "split_k": 8}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 64, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 32, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 64, 32], "work_group_sizes": [128, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 64, 64, 32], "work_group_sizes": [128, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4, "split_k": 8}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 64, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 32, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 64, 32], "work_group_sizes": [128, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 64, 64, 32], "work_group_sizes": [128, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4, "split_k": 8}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 64, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 32, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 64, 32], "work_group_sizes": [128, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 64, 64, 32], "work_group_sizes": [128, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4, "split_k": 8}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 64, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 32, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 64, 32], "work_group_sizes": [128, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 64, 64, 32], "work_group_sizes": [128, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4, "split_k": 8}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 64, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 32, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 64, 32], "work_group_sizes": [128, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 64, 64, 32], "work_group_sizes": [128, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4, "split_k": 8}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 64, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 32, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 64, 32], "work_group_sizes": [128, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 64, 64, 32], "work_group_sizes": [128, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4, "split_k": 8}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 64, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 32, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 64, 32], "work_group_sizes": [128, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 64, 64, 32], "work_group_sizes": [128, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4, "split_k": 8}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 64, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 32, 32, 16], "work_group_sizes": [64, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 32, 16], "work_group_sizes": [64, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [32, 64, 32], "work_group_sizes": [128, 1, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4}, {"work_group_tile_sizes": [1, 64, 64, 32], "work_group_sizes": [128, 2, 1], "pipeline": "GPU_TENSORCORE", "pipeline_depth": 4, "split_k": 8}, {"work_group_tile_sizes": [1, 32, 128], "work_group_sizes": [32, 1, 1], "pipeline": "GPU"}]}