@hoff97/tensor-js/dist/lib/onnx/nodes/upsample.js

PyTorch like deep learning inferrence library

var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
    function adopt(value) { return value instanceof P ? value : new P(function (resolve) { resolve(value); }); }
    return new (P || (P = Promise))(function (resolve, reject) {
        function fulfilled(value) { try { step(generator.next(value)); } catch (e) { reject(e); } }
        function rejected(value) { try { step(generator["throw"](value)); } catch (e) { reject(e); } }
        function step(result) { result.done ? resolve(result.value) : adopt(result.value).then(fulfilled, rejected); }
        step((generator = generator.apply(thisArg, _arguments || [])).next());
    });
};
import { CPUTensor } from '../../tensor/cpu/tensor';
import { toCPU } from '../../util/convert';
import { OnnxNode } from '../node';
export class UpsampleNode extends OnnxNode {
    constructor(attributes, inputs, outputs, constants, onnxVersion, mode) {
        super(attributes, inputs, outputs, constants, onnxVersion, mode);
        //@ts-ignore
        this.sampleMode = this.getAttributeString('mode');
        if (this.sampleMode !== 'nearest') {
            throw new Error('Upsampling only supported with nearest neighbor sampling');
        }
        if (this.onnxVersion < 9) {
            const scales = this.getAttributeFloats('scales');
            if (scales !== undefined && scales !== null) {
                this.scales = scales;
            }
            else {
                throw new Error(`Upsample node with onnx version ${this.onnxVersion} is missing scales attribute`);
            }
        }
    }
    getScales(scale) {
        return __awaiter(this, void 0, void 0, function* () {
            if (this.onnxVersion < 9) {
                return this.scales;
            }
            if (!(scale instanceof CPUTensor)) {
                console.warn('Scales tensor for upsample not on CPU, need to transfer!');
                scale = yield toCPU(scale);
            }
            const sc = scale;
            const scales = new Array(sc.size);
            for (let i = 0; i < sc.size; i++) {
                scales[i] = sc.get(i);
            }
            return scales;
        });
    }
    forward(inputs) {
        return __awaiter(this, void 0, void 0, function* () {
            const x = inputs[0];
            const scale = inputs[1];
            const scales = yield this.getScales(scale);
            return [x.upsample(scales)];
        });
    }
    getType() {
        return 'Upsample';
    }
    delete() { }
}
//# sourceMappingURL=upsample.js.map