@hoff97/tensor-js/dist/lib/ops/cpu/transpose.js

PyTorch like deep learning inferrence library

import { CPUTensor } from '../../tensor/cpu/tensor';
import { incrementIndex } from '../../util/shape';
export function transpose(x, permutation) {
    const rank = x.shape.length;
    const outputShape = new Array(rank);
    const reversePerm = new Array(rank);
    for (let i = 0; i < rank; i++) {
        outputShape[i] = x.shape[permutation[i]];
        reversePerm[permutation[i]] = i;
    }
    const result = new CPUTensor(outputShape, undefined, x.dtype);
    const resultStrides = result.strides;
    const mappedStrides = new Array(rank);
    for (let i = 0; i < rank; i++) {
        mappedStrides[i] = resultStrides[reversePerm[i]];
    }
    const index = new Array(rank).fill(0);
    for (let i = 0; i < x.size; i++) {
        let outIx = 0;
        for (let j = 0; j < rank; j++) {
            outIx += index[j] * mappedStrides[j];
        }
        result.set(outIx, x.get(i));
        incrementIndex(index, x.shape);
    }
    return result;
}
//# sourceMappingURL=transpose.js.map