@tensorflow-models/coco-ssd
Version:
Object detection model (coco-ssd) in TensorFlow.js
237 lines • 11.4 kB
JavaScript
"use strict";
var __awaiter = (this && this.__awaiter) || function (thisArg, _arguments, P, generator) {
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) : new P(function (resolve) { resolve(result.value); }).then(fulfilled, rejected); }
step((generator = generator.apply(thisArg, _arguments || [])).next());
});
};
var __generator = (this && this.__generator) || function (thisArg, body) {
var _ = { label: 0, sent: function() { if (t[0] & 1) throw t[1]; return t[1]; }, trys: [], ops: [] }, f, y, t, g;
return g = { next: verb(0), "throw": verb(1), "return": verb(2) }, typeof Symbol === "function" && (g[Symbol.iterator] = function() { return this; }), g;
function verb(n) { return function (v) { return step([n, v]); }; }
function step(op) {
if (f) throw new TypeError("Generator is already executing.");
while (_) try {
if (f = 1, y && (t = y[op[0] & 2 ? "return" : op[0] ? "throw" : "next"]) && !(t = t.call(y, op[1])).done) return t;
if (y = 0, t) op = [0, t.value];
switch (op[0]) {
case 0: case 1: t = op; break;
case 4: _.label++; return { value: op[1], done: false };
case 5: _.label++; y = op[1]; op = [0]; continue;
case 7: op = _.ops.pop(); _.trys.pop(); continue;
default:
if (!(t = _.trys, t = t.length > 0 && t[t.length - 1]) && (op[0] === 6 || op[0] === 2)) { _ = 0; continue; }
if (op[0] === 3 && (!t || (op[1] > t[0] && op[1] < t[3]))) { _.label = op[1]; break; }
if (op[0] === 6 && _.label < t[1]) { _.label = t[1]; t = op; break; }
if (t && _.label < t[2]) { _.label = t[2]; _.ops.push(op); break; }
if (t[2]) _.ops.pop();
_.trys.pop(); continue;
}
op = body.call(thisArg, _);
} catch (e) { op = [6, e]; y = 0; } finally { f = t = 0; }
if (op[0] & 5) throw op[1]; return { value: op[0] ? op[1] : void 0, done: true };
}
};
Object.defineProperty(exports, "__esModule", { value: true });
var tfc = require("@tensorflow/tfjs-core");
var compiled_api_1 = require("../data/compiled_api");
var operation_mapper_1 = require("../operations/operation_mapper");
var graph_executor_1 = require("./graph_executor");
exports.TFHUB_SEARCH_PARAM = '?tfjs-format=file';
exports.DEFAULT_MODEL_NAME = 'tensorflowjs_model.pb';
exports.DEFAULT_MANIFEST_NAME = 'weights_manifest.json';
var FrozenModel = (function () {
function FrozenModel(modelUrl, weightManifestUrl, requestOption, weightPrefix, onProgress) {
this.modelUrl = modelUrl;
this.weightManifestUrl = weightManifestUrl;
this.requestOption = requestOption;
this.weightPrefix = weightPrefix;
this.onProgress = onProgress;
this.version = 'n/a';
}
Object.defineProperty(FrozenModel.prototype, "modelVersion", {
get: function () {
return this.version;
},
enumerable: true,
configurable: true
});
Object.defineProperty(FrozenModel.prototype, "inputNodes", {
get: function () {
return this.executor.inputNodes;
},
enumerable: true,
configurable: true
});
Object.defineProperty(FrozenModel.prototype, "outputNodes", {
get: function () {
return this.executor.outputNodes;
},
enumerable: true,
configurable: true
});
Object.defineProperty(FrozenModel.prototype, "inputs", {
get: function () {
return this.executor.inputs;
},
enumerable: true,
configurable: true
});
Object.defineProperty(FrozenModel.prototype, "outputs", {
get: function () {
return this.executor.outputs;
},
enumerable: true,
configurable: true
});
Object.defineProperty(FrozenModel.prototype, "weights", {
get: function () {
return this.executor.weightMap;
},
enumerable: true,
configurable: true
});
FrozenModel.prototype.findIOHandler = function () {
var path = [this.modelUrl, this.weightManifestUrl];
if (this.requestOption || this.weightPrefix) {
this.handler = tfc.io.browserHTTPRequest(path, this.requestOption, this.weightPrefix, null, this.onProgress);
}
else {
var handlers = tfc.io.getLoadHandlers(path, this.onProgress);
if (handlers.length === 0) {
handlers.push(tfc.io.browserHTTPRequest(path, this.requestOption, this.weightPrefix, null, this.onProgress));
}
else if (handlers.length > 1) {
throw new Error("Found more than one (" + handlers.length + ") load handlers for " +
("URL '" + [path] + "'"));
}
this.handler = handlers[0];
}
};
FrozenModel.prototype.load = function () {
return __awaiter(this, void 0, void 0, function () {
var artifacts, graph, weightMap;
return __generator(this, function (_a) {
switch (_a.label) {
case 0:
this.findIOHandler();
if (this.handler.load == null) {
throw new Error('Cannot proceed with model loading because the IOHandler provided ' +
'does not have the `load` method implemented.');
}
return [4, this.handler.load()];
case 1:
artifacts = _a.sent();
graph = compiled_api_1.tensorflow.GraphDef.decode(new Uint8Array(artifacts.modelTopology));
this.version = graph.versions.producer + "." + graph.versions.minConsumer;
weightMap = tfc.io.decodeWeights(artifacts.weightData, artifacts.weightSpecs);
this.executor =
new graph_executor_1.GraphExecutor(operation_mapper_1.OperationMapper.Instance.transformGraph(graph));
this.executor.weightMap = this.convertTensorMapToTensorsMap(weightMap);
return [2, true];
}
});
});
};
FrozenModel.prototype.predict = function (inputs, config) {
return this.execute_(inputs, true, this.outputNodes);
};
FrozenModel.prototype.constructTensorMap = function (inputs) {
var inputArray = inputs instanceof tfc.Tensor ? [inputs] : inputs;
if (inputArray.length !== this.inputNodes.length) {
throw new Error('Input tensor count mismatch,' +
("the frozen model has " + this.inputNodes.length + " placeholders, ") +
("while there are " + inputArray.length + " input tensors."));
}
return this.inputNodes.reduce(function (map, inputName, i) {
map[inputName] = inputArray[i];
return map;
}, {});
};
FrozenModel.prototype.execute = function (inputs, outputs) {
return this.execute_(inputs, false, outputs);
};
FrozenModel.prototype.execute_ = function (inputs, strictInputCheck, outputs) {
if (strictInputCheck === void 0) { strictInputCheck = true; }
outputs = outputs || this.outputNodes;
if (inputs instanceof tfc.Tensor || Array.isArray(inputs)) {
inputs = this.constructTensorMap(inputs);
}
if (this.executor.isControlFlowModel || this.executor.isDynamicShapeModel) {
throw new Error('The model contains control flow or dynamic shape ops, ' +
'please use executeAsync method');
}
var result = this.executor.execute(this.convertTensorMapToTensorsMap(inputs), strictInputCheck, outputs);
var keys = Object.keys(result);
return (Array.isArray(outputs) && outputs.length > 1) ?
outputs.map(function (node) { return result[node]; }) :
result[keys[0]];
};
FrozenModel.prototype.executeAsync = function (inputs, outputs) {
return __awaiter(this, void 0, void 0, function () {
var result, keys;
return __generator(this, function (_a) {
switch (_a.label) {
case 0:
if (!(this.executor.isControlFlowModel ||
this.executor.isDynamicShapeModel)) {
throw new Error('The model does not contain control flow or dynamic shape ops, ' +
'please use execute method for better performance.');
}
outputs = outputs || this.outputNodes;
if (inputs instanceof tfc.Tensor || Array.isArray(inputs)) {
inputs = this.constructTensorMap(inputs);
}
return [4, this.executor.executeAsync(this.convertTensorMapToTensorsMap(inputs), outputs)];
case 1:
result = _a.sent();
keys = Object.keys(result);
return [2, Array.isArray(outputs) && outputs.length > 1 ?
outputs.map(function (node) { return result[node]; }) :
result[keys[0]]];
}
});
});
};
FrozenModel.prototype.convertTensorMapToTensorsMap = function (map) {
return Object.keys(map).reduce(function (newMap, key) {
newMap[key] = [map[key]];
return newMap;
}, {});
};
FrozenModel.prototype.dispose = function () {
this.executor.dispose();
};
return FrozenModel;
}());
exports.FrozenModel = FrozenModel;
function loadFrozenModel(modelUrl, weightsManifestUrl, requestOption, onProgress) {
return __awaiter(this, void 0, void 0, function () {
var model;
return __generator(this, function (_a) {
switch (_a.label) {
case 0:
model = new FrozenModel(modelUrl, weightsManifestUrl, requestOption, null, onProgress);
return [4, model.load()];
case 1:
_a.sent();
return [2, model];
}
});
});
}
exports.loadFrozenModel = loadFrozenModel;
function loadTfHubModule(tfhubModuleUrl, requestOption, onProgress) {
return __awaiter(this, void 0, void 0, function () {
return __generator(this, function (_a) {
if (!tfhubModuleUrl.endsWith('/')) {
tfhubModuleUrl = tfhubModuleUrl + '/';
}
return [2, loadFrozenModel("" + tfhubModuleUrl + exports.DEFAULT_MODEL_NAME + exports.TFHUB_SEARCH_PARAM, "" + tfhubModuleUrl + exports.DEFAULT_MANIFEST_NAME + exports.TFHUB_SEARCH_PARAM, requestOption, onProgress)];
});
});
}
exports.loadTfHubModule = loadTfHubModule;
//# sourceMappingURL=frozen_model.js.map