@azure-rest/ai-anomaly-detector 1.0.0-beta.1 → 1.0.0-beta.2

package/{dist-esm/src → dist/browser}/isUnexpected.js

@@ -1,5 +1,5 @@
 // Copyright (c) Microsoft Corporation.
-// Licensed under the MIT license.
+// Licensed under the MIT License.
 const responseMap = {
     "POST /timeseries/entire/detect": ["200"],
     "POST /timeseries/last/detect": ["200"],
@@ -14,7 +14,7 @@ const responseMap = {
 };
 export function isUnexpected(response) {
     const lroOriginal = response.headers["x-ms-original-url"];
-    const url = new URL(lroOriginal !== null && lroOriginal !== void 0 ? lroOriginal : response.request.url);
+    const url = new URL(lroOriginal ?? response.request.url);
     const method = response.request.method;
     let pathDetails = responseMap[`${method} ${url.pathname}`];
     if (!pathDetails) {
@@ -23,7 +23,6 @@ export function isUnexpected(response) {
     return !pathDetails.includes(response.status);
 }
 function getParametrizedPathSuccess(method, path) {
-    var _a, _b, _c, _d;
     const pathParts = path.split("/");
     // Traverse list to match the longest candidate
     // matchedLen: the length of candidate path
@@ -42,13 +41,13 @@ function getParametrizedPathSuccess(method, path) {
         // track if we have found a match to return the values found.
         let found = true;
         for (let i = candidateParts.length - 1, j = pathParts.length - 1; i >= 1 && j >= 1; i--, j--) {
-            if (((_a = candidateParts[i]) === null || _a === void 0 ? void 0 : _a.startsWith("{")) && ((_b = candidateParts[i]) === null || _b === void 0 ? void 0 : _b.indexOf("}")) !== -1) {
-                const start = candidateParts[i].indexOf("}") + 1, end = (_c = candidateParts[i]) === null || _c === void 0 ? void 0 : _c.length;
+            if (candidateParts[i]?.startsWith("{") && candidateParts[i]?.indexOf("}") !== -1) {
+                const start = candidateParts[i].indexOf("}") + 1, end = candidateParts[i]?.length;
                 // If the current part of the candidate is a "template" part
                 // Try to use the suffix of pattern to match the path
                 // {guid} ==> $
                 // {guid}:export ==> :export$
-                const isMatched = new RegExp(`${(_d = candidateParts[i]) === null || _d === void 0 ? void 0 : _d.slice(start, end)}`).test(pathParts[j] || "");
+                const isMatched = new RegExp(`${candidateParts[i]?.slice(start, end)}`).test(pathParts[j] || "");
                 if (!isMatched) {
                     found = false;
                     break;

package/dist/browser/isUnexpected.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"isUnexpected.js","sourceRoot":"","sources":["../../src/isUnexpected.ts"],"names":[],"mappings":"AAAA,uCAAuC;AACvC,kCAAkC;AAyBlC,MAAM,WAAW,GAA6B;IAC5C,gCAAgC,EAAE,CAAC,KAAK,CAAC;IACzC,8BAA8B,EAAE,CAAC,KAAK,CAAC;IACvC,qCAAqC,EAAE,CAAC,KAAK,CAAC;IAC9C,2CAA2C,EAAE,CAAC,KAAK,CAAC;IACpD,2BAA2B,EAAE,CAAC,KAAK,CAAC;IACpC,0BAA0B,EAAE,CAAC,KAAK,CAAC;IACnC,uCAAuC,EAAE,CAAC,KAAK,CAAC;IAChD,oCAAoC,EAAE,CAAC,KAAK,CAAC;IAC7C,kDAAkD,EAAE,CAAC,KAAK,CAAC;IAC3D,iDAAiD,EAAE,CAAC,KAAK,CAAC;CAC3D,CAAC;AAoCF,MAAM,UAAU,YAAY,CAC1B,QAoBgD;IAYhD,MAAM,WAAW,GAAG,QAAQ,CAAC,OAAO,CAAC,mBAAmB,CAAC,CAAC;IAC1D,MAAM,GAAG,GAAG,IAAI,GAAG,CAAC,WAAW,IAAI,QAAQ,CAAC,OAAO,CAAC,GAAG,CAAC,CAAC;IACzD,MAAM,MAAM,GAAG,QAAQ,CAAC,OAAO,CAAC,MAAM,CAAC;IACvC,IAAI,WAAW,GAAG,WAAW,CAAC,GAAG,MAAM,IAAI,GAAG,CAAC,QAAQ,EAAE,CAAC,CAAC;IAC3D,IAAI,CAAC,WAAW,EAAE,CAAC;QACjB,WAAW,GAAG,0BAA0B,CAAC,MAAM,EAAE,GAAG,CAAC,QAAQ,CAAC,CAAC;IACjE,CAAC;IACD,OAAO,CAAC,WAAW,CAAC,QAAQ,CAAC,QAAQ,CAAC,MAAM,CAAC,CAAC;AAChD,CAAC;AAED,SAAS,0BAA0B,CAAC,MAAc,EAAE,IAAY;IAC9D,MAAM,SAAS,GAAG,IAAI,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC;IAElC,+CAA+C;IAC/C,2CAA2C;IAC3C,8CAA8C;IAC9C,IAAI,UAAU,GAAG,CAAC,CAAC,EACjB,YAAY,GAAa,EAAE,CAAC;IAE9B,0CAA0C;IAC1C,KAAK,MAAM,CAAC,GAAG,EAAE,KAAK,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,WAAW,CAAC,EAAE,CAAC;QACvD,0DAA0D;QAC1D,gBAAgB;QAChB,IAAI,CAAC,GAAG,CAAC,UAAU,CAAC,MAAM,CAAC,EAAE,CAAC;YAC5B,SAAS;QACX,CAAC;QACD,MAAM,aAAa,GAAG,iBAAiB,CAAC,GAAG,CAAC,CAAC;QAC7C,gCAAgC;QAChC,MAAM,cAAc,GAAG,aAAa,CAAC,KAAK,CAAC,GAAG,CAAC,CAAC;QAEhD,6DAA6D;QAC7D,IAAI,KAAK,GAAG,IAAI,CAAC;QACjB,KAAK,IAAI,CAAC,GAAG,cAAc,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC,GAAG,SAAS,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC,IAAI,CAAC,IAAI,CAAC,IAAI,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,EAAE,EAAE,CAAC;YAC7F,IAAI,cAAc,CAAC,CAAC,CAAC,EAAE,UAAU,CAAC,GAAG,CAAC,IAAI,cAAc,CAAC,CAAC,CAAC,EAAE,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,CAAC,EAAE,CAAC;gBACjF,MAAM,KAAK,GAAG,cAAc,CAAC,CAAC,CAAE,CAAC,OAAO,CAAC,GAAG,CAAC,GAAG,CAAC,EAC/C,GAAG,GAAG,cAAc,CAAC,CAAC,CAAC,EAAE,MAAM,CAAC;gBAClC,4DAA4D;gBAC5D,qDAAqD;gBACrD,eAAe;gBACf,6BAA6B;gBAC7B,MAAM,SAAS,GAAG,IAAI,MAAM,CAAC,GAAG,cAAc,CAAC,CAAC,CAAC,EAAE,KAAK,CAAC,KAAK,EAAE,GAAG,CAAC,EAAE,CAAC,CAAC,IAAI,CAC1E,SAAS,CAAC,CAAC,CAAC,IAAI,EAAE,CACnB,CAAC;gBAEF,IAAI,CAAC,SAAS,EAAE,CAAC;oBACf,KAAK,GAAG,KAAK,CAAC;oBACd,MAAM;gBACR,CAAC;gBACD,SAAS;YACX,CAAC;YAED,8CAA8C;YAC9C,wDAAwD;YACxD,2CAA2C;YAC3C,IAAI,cAAc,CAAC,CAAC,CAAC,KAAK,SAAS,CAAC,CAAC,CAAC,EAAE,CAAC;gBACvC,KAAK,GAAG,KAAK,CAAC;gBACd,MAAM;YACR,CAAC;QACH,CAAC;QAED,qDAAqD;QACrD,sEAAsE;QACtE,IAAI,KAAK,IAAI,aAAa,CAAC,MAAM,GAAG,UAAU,EAAE,CAAC;YAC/C,UAAU,GAAG,aAAa,CAAC,MAAM,CAAC;YAClC,YAAY,GAAG,KAAK,CAAC;QACvB,CAAC;IACH,CAAC;IAED,OAAO,YAAY,CAAC;AACtB,CAAC;AAED,SAAS,iBAAiB,CAAC,MAAc;IACvC,MAAM,SAAS,GAAG,MAAM,CAAC,OAAO,CAAC,GAAG,CAAC,CAAC;IACtC,OAAO,MAAM,CAAC,KAAK,CAAC,SAAS,CAAC,CAAC;AACjC,CAAC","sourcesContent":["// Copyright (c) Microsoft Corporation.\n// Licensed under the MIT License.\n\nimport type {\n  DetectUnivariateEntireSeries200Response,\n  DetectUnivariateEntireSeriesDefaultResponse,\n  DetectUnivariateLastPoint200Response,\n  DetectUnivariateLastPointDefaultResponse,\n  DetectUnivariateChangePoint200Response,\n  DetectUnivariateChangePointDefaultResponse,\n  GetMultivariateBatchDetectionResult200Response,\n  GetMultivariateBatchDetectionResultDefaultResponse,\n  TrainMultivariateModel201Response,\n  TrainMultivariateModelDefaultResponse,\n  ListMultivariateModels200Response,\n  ListMultivariateModelsDefaultResponse,\n  DeleteMultivariateModel204Response,\n  DeleteMultivariateModelDefaultResponse,\n  GetMultivariateModel200Response,\n  GetMultivariateModelDefaultResponse,\n  DetectMultivariateBatchAnomaly202Response,\n  DetectMultivariateBatchAnomalyDefaultResponse,\n  DetectMultivariateLastAnomaly200Response,\n  DetectMultivariateLastAnomalyDefaultResponse,\n} from \"./responses.js\";\n\nconst responseMap: Record<string, string[]> = {\n  \"POST /timeseries/entire/detect\": [\"200\"],\n  \"POST /timeseries/last/detect\": [\"200\"],\n  \"POST /timeseries/changepoint/detect\": [\"200\"],\n  \"GET /multivariate/detect-batch/{resultId}\": [\"200\"],\n  \"POST /multivariate/models\": [\"201\"],\n  \"GET /multivariate/models\": [\"200\"],\n  \"DELETE /multivariate/models/{modelId}\": [\"204\"],\n  \"GET /multivariate/models/{modelId}\": [\"200\"],\n  \"POST /multivariate/models/{modelId}:detect-batch\": [\"202\"],\n  \"POST /multivariate/models/{modelId}:detect-last\": [\"200\"],\n};\n\nexport function isUnexpected(\n  response: DetectUnivariateEntireSeries200Response | DetectUnivariateEntireSeriesDefaultResponse,\n): response is DetectUnivariateEntireSeriesDefaultResponse;\nexport function isUnexpected(\n  response: DetectUnivariateLastPoint200Response | DetectUnivariateLastPointDefaultResponse,\n): response is DetectUnivariateLastPointDefaultResponse;\nexport function isUnexpected(\n  response: DetectUnivariateChangePoint200Response | DetectUnivariateChangePointDefaultResponse,\n): response is DetectUnivariateChangePointDefaultResponse;\nexport function isUnexpected(\n  response:\n    | GetMultivariateBatchDetectionResult200Response\n    | GetMultivariateBatchDetectionResultDefaultResponse,\n): response is GetMultivariateBatchDetectionResultDefaultResponse;\nexport function isUnexpected(\n  response: TrainMultivariateModel201Response | TrainMultivariateModelDefaultResponse,\n): response is TrainMultivariateModelDefaultResponse;\nexport function isUnexpected(\n  response: ListMultivariateModels200Response | ListMultivariateModelsDefaultResponse,\n): response is ListMultivariateModelsDefaultResponse;\nexport function isUnexpected(\n  response: DeleteMultivariateModel204Response | DeleteMultivariateModelDefaultResponse,\n): response is DeleteMultivariateModelDefaultResponse;\nexport function isUnexpected(\n  response: GetMultivariateModel200Response | GetMultivariateModelDefaultResponse,\n): response is GetMultivariateModelDefaultResponse;\nexport function isUnexpected(\n  response:\n    | DetectMultivariateBatchAnomaly202Response\n    | DetectMultivariateBatchAnomalyDefaultResponse,\n): response is DetectMultivariateBatchAnomalyDefaultResponse;\nexport function isUnexpected(\n  response: DetectMultivariateLastAnomaly200Response | DetectMultivariateLastAnomalyDefaultResponse,\n): response is DetectMultivariateLastAnomalyDefaultResponse;\nexport function isUnexpected(\n  response:\n    | DetectUnivariateEntireSeries200Response\n    | DetectUnivariateEntireSeriesDefaultResponse\n    | DetectUnivariateLastPoint200Response\n    | DetectUnivariateLastPointDefaultResponse\n    | DetectUnivariateChangePoint200Response\n    | DetectUnivariateChangePointDefaultResponse\n    | GetMultivariateBatchDetectionResult200Response\n    | GetMultivariateBatchDetectionResultDefaultResponse\n    | TrainMultivariateModel201Response\n    | TrainMultivariateModelDefaultResponse\n    | ListMultivariateModels200Response\n    | ListMultivariateModelsDefaultResponse\n    | DeleteMultivariateModel204Response\n    | DeleteMultivariateModelDefaultResponse\n    | GetMultivariateModel200Response\n    | GetMultivariateModelDefaultResponse\n    | DetectMultivariateBatchAnomaly202Response\n    | DetectMultivariateBatchAnomalyDefaultResponse\n    | DetectMultivariateLastAnomaly200Response\n    | DetectMultivariateLastAnomalyDefaultResponse,\n): response is\n  | DetectUnivariateEntireSeriesDefaultResponse\n  | DetectUnivariateLastPointDefaultResponse\n  | DetectUnivariateChangePointDefaultResponse\n  | GetMultivariateBatchDetectionResultDefaultResponse\n  | TrainMultivariateModelDefaultResponse\n  | ListMultivariateModelsDefaultResponse\n  | DeleteMultivariateModelDefaultResponse\n  | GetMultivariateModelDefaultResponse\n  | DetectMultivariateBatchAnomalyDefaultResponse\n  | DetectMultivariateLastAnomalyDefaultResponse {\n  const lroOriginal = response.headers[\"x-ms-original-url\"];\n  const url = new URL(lroOriginal ?? response.request.url);\n  const method = response.request.method;\n  let pathDetails = responseMap[`${method} ${url.pathname}`];\n  if (!pathDetails) {\n    pathDetails = getParametrizedPathSuccess(method, url.pathname);\n  }\n  return !pathDetails.includes(response.status);\n}\n\nfunction getParametrizedPathSuccess(method: string, path: string): string[] {\n  const pathParts = path.split(\"/\");\n\n  // Traverse list to match the longest candidate\n  // matchedLen: the length of candidate path\n  // matchedValue: the matched status code array\n  let matchedLen = -1,\n    matchedValue: string[] = [];\n\n  // Iterate the responseMap to find a match\n  for (const [key, value] of Object.entries(responseMap)) {\n    // Extracting the path from the map key which is in format\n    // GET /path/foo\n    if (!key.startsWith(method)) {\n      continue;\n    }\n    const candidatePath = getPathFromMapKey(key);\n    // Get each part of the url path\n    const candidateParts = candidatePath.split(\"/\");\n\n    // track if we have found a match to return the values found.\n    let found = true;\n    for (let i = candidateParts.length - 1, j = pathParts.length - 1; i >= 1 && j >= 1; i--, j--) {\n      if (candidateParts[i]?.startsWith(\"{\") && candidateParts[i]?.indexOf(\"}\") !== -1) {\n        const start = candidateParts[i]!.indexOf(\"}\") + 1,\n          end = candidateParts[i]?.length;\n        // If the current part of the candidate is a \"template\" part\n        // Try to use the suffix of pattern to match the path\n        // {guid} ==> $\n        // {guid}:export ==> :export$\n        const isMatched = new RegExp(`${candidateParts[i]?.slice(start, end)}`).test(\n          pathParts[j] || \"\",\n        );\n\n        if (!isMatched) {\n          found = false;\n          break;\n        }\n        continue;\n      }\n\n      // If the candidate part is not a template and\n      // the parts don't match mark the candidate as not found\n      // to move on with the next candidate path.\n      if (candidateParts[i] !== pathParts[j]) {\n        found = false;\n        break;\n      }\n    }\n\n    // We finished evaluating the current candidate parts\n    // Update the matched value if and only if we found the longer pattern\n    if (found && candidatePath.length > matchedLen) {\n      matchedLen = candidatePath.length;\n      matchedValue = value;\n    }\n  }\n\n  return matchedValue;\n}\n\nfunction getPathFromMapKey(mapKey: string): string {\n  const pathStart = mapKey.indexOf(\"/\");\n  return mapKey.slice(pathStart);\n}\n"]}

package/dist/browser/models.d.ts

@@ -0,0 +1,252 @@
+/** The request of entire or last anomaly detection. */
+export interface UnivariateDetectionOptions {
+    /**
+     * Time series data points. Points should be sorted by timestamp in ascending
+     * order to match the anomaly detection result. If the data is not sorted
+     * correctly or there is duplicated timestamp, the API will not work. In such
+     * case, an error message will be returned.
+     */
+    series: Array<TimeSeriesPoint>;
+    /**
+     * Optional argument, can be one of yearly, monthly, weekly, daily, hourly,
+     * minutely, secondly, microsecond or none. If granularity is not present, it will
+     * be none by default. If granularity is none, the timestamp property in time
+     * series point can be absent.
+     */
+    granularity?: "yearly" | "monthly" | "weekly" | "daily" | "hourly" | "minutely" | "secondly" | "microsecond" | "none";
+    /**
+     * Custom Interval is used to set non-standard time interval, for example, if the
+     * series is 5 minutes, request can be set as {"granularity":"minutely",
+     * "customInterval":5}.
+     */
+    customInterval?: number;
+    /**
+     * Optional argument, periodic value of a time series. If the value is null or
+     * does not present, the API will determine the period automatically.
+     */
+    period?: number;
+    /** Optional argument, advanced model parameter, max anomaly ratio in a time series. */
+    maxAnomalyRatio?: number;
+    /**
+     * Optional argument, advanced model parameter, between 0-99, the lower the value
+     * is, the larger the margin value will be which means less anomalies will be
+     * accepted.
+     */
+    sensitivity?: number;
+    /**
+     * Used to specify how to deal with missing values in the input series, it's used
+     * when granularity is not "none".
+     *
+     * Possible values: auto, previous, linear, fixed, zero, notFill
+     */
+    imputeMode?: string;
+    /**
+     * Used to specify the value to fill, it's used when granularity is not "none"
+     * and imputeMode is "fixed".
+     */
+    imputeFixedValue?: number;
+}
+/** The definition of input timeseries points. */
+export interface TimeSeriesPoint {
+    /** Optional argument, timestamp of a data point (ISO8601 format). */
+    timestamp?: Date | string;
+    /** The measurement of that point, should be float. */
+    value: number;
+}
+/** The request of change point detection. */
+export interface UnivariateChangePointDetectionOptions {
+    /**
+     * Time series data points. Points should be sorted by timestamp in ascending
+     * order to match the change point detection result.
+     */
+    series: Array<TimeSeriesPoint>;
+    /**
+     * Can only be one of yearly, monthly, weekly, daily, hourly, minutely or
+     * secondly. Granularity is used for verify whether input series is valid.
+     */
+    granularity: "yearly" | "monthly" | "weekly" | "daily" | "hourly" | "minutely" | "secondly" | "microsecond" | "none";
+    /**
+     * Custom Interval is used to set non-standard time interval, for example, if the
+     * series is 5 minutes, request can be set as {"granularity":"minutely",
+     * "customInterval":5}.
+     */
+    customInterval?: number;
+    /**
+     * Optional argument, periodic value of a time series. If the value is null or
+     * does not present, the API will determine the period automatically.
+     */
+    period?: number;
+    /**
+     * Optional argument, advanced model parameter, a default stableTrendWindow will
+     * be used in detection.
+     */
+    stableTrendWindow?: number;
+    /**
+     * Optional argument, advanced model parameter, between 0.0-1.0, the lower the
+     * value is, the larger the trend error will be which means less change point will
+     * be accepted.
+     */
+    threshold?: number;
+}
+/** ErrorResponse contains code and message that shows the error information. */
+export interface ErrorResponse {
+    /** The error code. */
+    code: string;
+    /** The message explaining the error reported by the service. */
+    message: string;
+}
+/** Variable Status. */
+export interface VariableState {
+    /** Variable name in variable states. */
+    variable?: string;
+    /** Proportion of missing values that need to be filled by fillNAMethod. */
+    filledNARatio?: number;
+    /** Number of effective data points before applying fillNAMethod. */
+    effectiveCount?: number;
+    /** First valid timestamp with value of input data. */
+    firstTimestamp?: Date | string;
+    /** Last valid timestamp with value of input data. */
+    lastTimestamp?: Date | string;
+}
+/**
+ * Detection request for batch inference. This is an asynchronous inference which
+ * will need another API to get detection results.
+ */
+export interface MultivariateBatchDetectionOptions {
+    /**
+     * Source link to the input data to indicate an accessible Azure storage Uri,
+     * either pointed to an Azure blob storage folder, or pointed to a CSV file in
+     * Azure blob storage based on you data schema selection. The data schema should
+     * be exactly the same with those used in the training phase.
+     */
+    dataSource: string;
+    /**
+     * An optional field, which is used to specify the number of top contributed
+     * variables for one anomalous timestamp in the response. The default number is
+     * 10.
+     */
+    topContributorCount: number;
+    /**
+     * A required field, indicating the start time of data for detection, which should
+     * be date-time of ISO 8601 format.
+     */
+    startTime: Date | string;
+    /**
+     * A required field, indicating the end time of data for detection, which should
+     * be date-time of ISO 8601 format.
+     */
+    endTime: Date | string;
+}
+/**
+ * Training result of a model including its status, errors and diagnostics
+ * information.
+ */
+export interface ModelInfo {
+    /**
+     * Source link to the input data to indicate an accessible Azure storage Uri,
+     * either pointed to an Azure blob storage folder, or pointed to a CSV file in
+     * Azure blob storage based on you data schema selection.
+     */
+    dataSource: string;
+    /**
+     * Data schema of input data source: OneTable or MultiTable. The default
+     * DataSchema is OneTable.
+     *
+     * Possible values: OneTable, MultiTable
+     */
+    dataSchema?: string;
+    /**
+     * A required field, indicating the start time of training data, which should be
+     * date-time of ISO 8601 format.
+     */
+    startTime: Date | string;
+    /**
+     * A required field, indicating the end time of training data, which should be
+     * date-time of ISO 8601 format.
+     */
+    endTime: Date | string;
+    /**
+     * An optional field. The display name of the model whose maximum length is 24
+     * characters.
+     */
+    displayName?: string;
+    /**
+     * An optional field, indicating how many previous timestamps will be used to
+     * detect whether the timestamp is anomaly or not.
+     */
+    slidingWindow?: number;
+    /** An optional field, indicating the manner to align multiple variables. */
+    alignPolicy?: AlignPolicy;
+    /** Model status. One of CREATED, RUNNING, READY, and FAILED. */
+    status?: "CREATED" | "RUNNING" | "READY" | "FAILED";
+    /** Diagnostics information to help inspect the states of model or variable. */
+    diagnosticsInfo?: DiagnosticsInfo;
+}
+/** An optional field, indicating the manner to align multiple variables. */
+export interface AlignPolicy {
+    /**
+     * An optional field, indicating how to align different variables to the same
+     * time-range. Either Inner or Outer.
+     */
+    alignMode?: "Inner" | "Outer";
+    /**
+     * An optional field, indicating how missing values will be filled. One of
+     * Previous, Subsequent, Linear, Zero, Fixed.
+     *
+     * Possible values: Previous, Subsequent, Linear, Zero, Fixed
+     */
+    fillNAMethod?: string;
+    /** An optional field. Required when fillNAMethod is Fixed. */
+    paddingValue?: number;
+}
+/** Diagnostics information to help inspect the states of model or variable. */
+export interface DiagnosticsInfo {
+    /** Model status. */
+    modelState?: ModelState;
+    /** Variable Status. */
+    variableStates?: Array<VariableState>;
+}
+/** Model status. */
+export interface ModelState {
+    /**
+     * This indicates the number of passes of the entire training dataset the
+     * algorithm has completed.
+     */
+    epochIds?: number[];
+    /**
+     * List of metrics used to assess how the model fits the training data for each
+     * epoch.
+     */
+    trainLosses?: number[];
+    /**
+     * List of metrics used to assess how the model fits the validation set for each
+     * epoch.
+     */
+    validationLosses?: number[];
+    /** Latency for each epoch. */
+    latenciesInSeconds?: number[];
+}
+/** Request of last detection. */
+export interface MultivariateLastDetectionOptions {
+    /**
+     * This contains the inference data, including the name, timestamps(ISO 8601) and
+     * values of variables.
+     */
+    variables: Array<VariableValues>;
+    /**
+     * An optional field, which is used to specify the number of top contributed
+     * variables for one anomalous timestamp in the response. The default number is
+     * 10.
+     */
+    topContributorCount: number;
+}
+/** Variable values. */
+export interface VariableValues {
+    /** Variable name of last detection request. */
+    variable: string;
+    /** Timestamps of last detection request */
+    timestamps: string[];
+    /** Values of variables. */
+    values: number[];
+}
+//# sourceMappingURL=models.d.ts.map

package/dist/browser/models.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"models.d.ts","sourceRoot":"","sources":["../../src/models.ts"],"names":[],"mappings":"AAGA,uDAAuD;AACvD,MAAM,WAAW,0BAA0B;IACzC;;;;;OAKG;IACH,MAAM,EAAE,KAAK,CAAC,eAAe,CAAC,CAAC;IAC/B;;;;;OAKG;IACH,WAAW,CAAC,EACR,QAAQ,GACR,SAAS,GACT,QAAQ,GACR,OAAO,GACP,QAAQ,GACR,UAAU,GACV,UAAU,GACV,aAAa,GACb,MAAM,CAAC;IACX;;;;OAIG;IACH,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB;;;OAGG;IACH,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,uFAAuF;IACvF,eAAe,CAAC,EAAE,MAAM,CAAC;IACzB;;;;OAIG;IACH,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB;;;;;OAKG;IACH,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB;;;OAGG;IACH,gBAAgB,CAAC,EAAE,MAAM,CAAC;CAC3B;AAED,iDAAiD;AACjD,MAAM,WAAW,eAAe;IAC9B,qEAAqE;IACrE,SAAS,CAAC,EAAE,IAAI,GAAG,MAAM,CAAC;IAC1B,sDAAsD;IACtD,KAAK,EAAE,MAAM,CAAC;CACf;AAED,6CAA6C;AAC7C,MAAM,WAAW,qCAAqC;IACpD;;;OAGG;IACH,MAAM,EAAE,KAAK,CAAC,eAAe,CAAC,CAAC;IAC/B;;;OAGG;IACH,WAAW,EACP,QAAQ,GACR,SAAS,GACT,QAAQ,GACR,OAAO,GACP,QAAQ,GACR,UAAU,GACV,UAAU,GACV,aAAa,GACb,MAAM,CAAC;IACX;;;;OAIG;IACH,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB;;;OAGG;IACH,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB;;;OAGG;IACH,iBAAiB,CAAC,EAAE,MAAM,CAAC;IAC3B;;;;OAIG;IACH,SAAS,CAAC,EAAE,MAAM,CAAC;CACpB;AAED,gFAAgF;AAChF,MAAM,WAAW,aAAa;IAC5B,sBAAsB;IACtB,IAAI,EAAE,MAAM,CAAC;IACb,gEAAgE;IAChE,OAAO,EAAE,MAAM,CAAC;CACjB;AAED,uBAAuB;AACvB,MAAM,WAAW,aAAa;IAC5B,wCAAwC;IACxC,QAAQ,CAAC,EAAE,MAAM,CAAC;IAClB,2EAA2E;IAC3E,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB,oEAAoE;IACpE,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB,sDAAsD;IACtD,cAAc,CAAC,EAAE,IAAI,GAAG,MAAM,CAAC;IAC/B,qDAAqD;IACrD,aAAa,CAAC,EAAE,IAAI,GAAG,MAAM,CAAC;CAC/B;AAED;;;GAGG;AACH,MAAM,WAAW,iCAAiC;IAChD;;;;;OAKG;IACH,UAAU,EAAE,MAAM,CAAC;IACnB;;;;OAIG;IACH,mBAAmB,EAAE,MAAM,CAAC;IAC5B;;;OAGG;IACH,SAAS,EAAE,IAAI,GAAG,MAAM,CAAC;IACzB;;;OAGG;IACH,OAAO,EAAE,IAAI,GAAG,MAAM,CAAC;CACxB;AAED;;;GAGG;AACH,MAAM,WAAW,SAAS;IACxB;;;;OAIG;IACH,UAAU,EAAE,MAAM,CAAC;IACnB;;;;;OAKG;IACH,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB;;;OAGG;IACH,SAAS,EAAE,IAAI,GAAG,MAAM,CAAC;IACzB;;;OAGG;IACH,OAAO,EAAE,IAAI,GAAG,MAAM,CAAC;IACvB;;;OAGG;IACH,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB;;;OAGG;IACH,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB,4EAA4E;IAC5E,WAAW,CAAC,EAAE,WAAW,CAAC;IAC1B,gEAAgE;IAChE,MAAM,CAAC,EAAE,SAAS,GAAG,SAAS,GAAG,OAAO,GAAG,QAAQ,CAAC;IACpD,+EAA+E;IAC/E,eAAe,CAAC,EAAE,eAAe,CAAC;CACnC;AAED,4EAA4E;AAC5E,MAAM,WAAW,WAAW;IAC1B;;;OAGG;IACH,SAAS,CAAC,EAAE,OAAO,GAAG,OAAO,CAAC;IAC9B;;;;;OAKG;IACH,YAAY,CAAC,EAAE,MAAM,CAAC;IACtB,8DAA8D;IAC9D,YAAY,CAAC,EAAE,MAAM,CAAC;CACvB;AAED,+EAA+E;AAC/E,MAAM,WAAW,eAAe;IAC9B,oBAAoB;IACpB,UAAU,CAAC,EAAE,UAAU,CAAC;IACxB,uBAAuB;IACvB,cAAc,CAAC,EAAE,KAAK,CAAC,aAAa,CAAC,CAAC;CACvC;AAED,oBAAoB;AACpB,MAAM,WAAW,UAAU;IACzB;;;OAGG;IACH,QAAQ,CAAC,EAAE,MAAM,EAAE,CAAC;IACpB;;;OAGG;IACH,WAAW,CAAC,EAAE,MAAM,EAAE,CAAC;IACvB;;;OAGG;IACH,gBAAgB,CAAC,EAAE,MAAM,EAAE,CAAC;IAC5B,8BAA8B;IAC9B,kBAAkB,CAAC,EAAE,MAAM,EAAE,CAAC;CAC/B;AAED,iCAAiC;AACjC,MAAM,WAAW,gCAAgC;IAC/C;;;OAGG;IACH,SAAS,EAAE,KAAK,CAAC,cAAc,CAAC,CAAC;IACjC;;;;OAIG;IACH,mBAAmB,EAAE,MAAM,CAAC;CAC7B;AAED,uBAAuB;AACvB,MAAM,WAAW,cAAc;IAC7B,+CAA+C;IAC/C,QAAQ,EAAE,MAAM,CAAC;IACjB,2CAA2C;IAC3C,UAAU,EAAE,MAAM,EAAE,CAAC;IACrB,2BAA2B;IAC3B,MAAM,EAAE,MAAM,EAAE,CAAC;CAClB"}

package/dist/browser/models.js

@@ -0,0 +1,4 @@
+// Copyright (c) Microsoft Corporation.
+// Licensed under the MIT License.
+export {};
+//# sourceMappingURL=models.js.map

package/{dist-esm/src → dist/browser}/models.js.map

@@ -1 +1 @@
-{"version":3,"file":"models.js","sourceRoot":"","sources":["../../src/models.ts"],"names":[],"mappings":"AAAA,uCAAuC;AACvC,kCAAkC","sourcesContent":["// Copyright (c) Microsoft Corporation.\n// Licensed under the MIT license.\n\n/** The request of entire or last anomaly detection. */\nexport interface UnivariateDetectionOptions {\n  /**\n   * Time series data points. Points should be sorted by timestamp in ascending\n   * order to match the anomaly detection result. If the data is not sorted\n   * correctly or there is duplicated timestamp, the API will not work. In such\n   * case, an error message will be returned.\n   */\n  series: Array<TimeSeriesPoint>;\n  /**\n   * Optional argument, can be one of yearly, monthly, weekly, daily, hourly,\n   * minutely, secondly, microsecond or none. If granularity is not present, it will\n   * be none by default. If granularity is none, the timestamp property in time\n   * series point can be absent.\n   */\n  granularity?:\n    | \"yearly\"\n    | \"monthly\"\n    | \"weekly\"\n    | \"daily\"\n    | \"hourly\"\n    | \"minutely\"\n    | \"secondly\"\n    | \"microsecond\"\n    | \"none\";\n  /**\n   * Custom Interval is used to set non-standard time interval, for example, if the\n   * series is 5 minutes, request can be set as {\"granularity\":\"minutely\",\n   * \"customInterval\":5}.\n   */\n  customInterval?: number;\n  /**\n   * Optional argument, periodic value of a time series. If the value is null or\n   * does not present, the API will determine the period automatically.\n   */\n  period?: number;\n  /** Optional argument, advanced model parameter, max anomaly ratio in a time series. */\n  maxAnomalyRatio?: number;\n  /**\n   * Optional argument, advanced model parameter, between 0-99, the lower the value\n   * is, the larger the margin value will be which means less anomalies will be\n   * accepted.\n   */\n  sensitivity?: number;\n  /**\n   * Used to specify how to deal with missing values in the input series, it's used\n   * when granularity is not \"none\".\n   *\n   * Possible values: auto, previous, linear, fixed, zero, notFill\n   */\n  imputeMode?: string;\n  /**\n   * Used to specify the value to fill, it's used when granularity is not \"none\"\n   * and imputeMode is \"fixed\".\n   */\n  imputeFixedValue?: number;\n}\n\n/** The definition of input timeseries points. */\nexport interface TimeSeriesPoint {\n  /** Optional argument, timestamp of a data point (ISO8601 format). */\n  timestamp?: Date | string;\n  /** The measurement of that point, should be float. */\n  value: number;\n}\n\n/** The request of change point detection. */\nexport interface UnivariateChangePointDetectionOptions {\n  /**\n   * Time series data points. Points should be sorted by timestamp in ascending\n   * order to match the change point detection result.\n   */\n  series: Array<TimeSeriesPoint>;\n  /**\n   * Can only be one of yearly, monthly, weekly, daily, hourly, minutely or\n   * secondly. Granularity is used for verify whether input series is valid.\n   */\n  granularity:\n    | \"yearly\"\n    | \"monthly\"\n    | \"weekly\"\n    | \"daily\"\n    | \"hourly\"\n    | \"minutely\"\n    | \"secondly\"\n    | \"microsecond\"\n    | \"none\";\n  /**\n   * Custom Interval is used to set non-standard time interval, for example, if the\n   * series is 5 minutes, request can be set as {\"granularity\":\"minutely\",\n   * \"customInterval\":5}.\n   */\n  customInterval?: number;\n  /**\n   * Optional argument, periodic value of a time series. If the value is null or\n   * does not present, the API will determine the period automatically.\n   */\n  period?: number;\n  /**\n   * Optional argument, advanced model parameter, a default stableTrendWindow will\n   * be used in detection.\n   */\n  stableTrendWindow?: number;\n  /**\n   * Optional argument, advanced model parameter, between 0.0-1.0, the lower the\n   * value is, the larger the trend error will be which means less change point will\n   * be accepted.\n   */\n  threshold?: number;\n}\n\n/** ErrorResponse contains code and message that shows the error information. */\nexport interface ErrorResponse {\n  /** The error code. */\n  code: string;\n  /** The message explaining the error reported by the service. */\n  message: string;\n}\n\n/** Variable Status. */\nexport interface VariableState {\n  /** Variable name in variable states. */\n  variable?: string;\n  /** Proportion of missing values that need to be filled by fillNAMethod. */\n  filledNARatio?: number;\n  /** Number of effective data points before applying fillNAMethod. */\n  effectiveCount?: number;\n  /** First valid timestamp with value of input data. */\n  firstTimestamp?: Date | string;\n  /** Last valid timestamp with value of input data. */\n  lastTimestamp?: Date | string;\n}\n\n/**\n * Detection request for batch inference. This is an asynchronous inference which\n * will need another API to get detection results.\n */\nexport interface MultivariateBatchDetectionOptions {\n  /**\n   * Source link to the input data to indicate an accessible Azure storage Uri,\n   * either pointed to an Azure blob storage folder, or pointed to a CSV file in\n   * Azure blob storage based on you data schema selection. The data schema should\n   * be exactly the same with those used in the training phase.\n   */\n  dataSource: string;\n  /**\n   * An optional field, which is used to specify the number of top contributed\n   * variables for one anomalous timestamp in the response. The default number is\n   * 10.\n   */\n  topContributorCount: number;\n  /**\n   * A required field, indicating the start time of data for detection, which should\n   * be date-time of ISO 8601 format.\n   */\n  startTime: Date | string;\n  /**\n   * A required field, indicating the end time of data for detection, which should\n   * be date-time of ISO 8601 format.\n   */\n  endTime: Date | string;\n}\n\n/**\n * Training result of a model including its status, errors and diagnostics\n * information.\n */\nexport interface ModelInfo {\n  /**\n   * Source link to the input data to indicate an accessible Azure storage Uri,\n   * either pointed to an Azure blob storage folder, or pointed to a CSV file in\n   * Azure blob storage based on you data schema selection.\n   */\n  dataSource: string;\n  /**\n   * Data schema of input data source: OneTable or MultiTable. The default\n   * DataSchema is OneTable.\n   *\n   * Possible values: OneTable, MultiTable\n   */\n  dataSchema?: string;\n  /**\n   * A required field, indicating the start time of training data, which should be\n   * date-time of ISO 8601 format.\n   */\n  startTime: Date | string;\n  /**\n   * A required field, indicating the end time of training data, which should be\n   * date-time of ISO 8601 format.\n   */\n  endTime: Date | string;\n  /**\n   * An optional field. The display name of the model whose maximum length is 24\n   * characters.\n   */\n  displayName?: string;\n  /**\n   * An optional field, indicating how many previous timestamps will be used to\n   * detect whether the timestamp is anomaly or not.\n   */\n  slidingWindow?: number;\n  /** An optional field, indicating the manner to align multiple variables. */\n  alignPolicy?: AlignPolicy;\n  /** Model status. One of CREATED, RUNNING, READY, and FAILED. */\n  status?: \"CREATED\" | \"RUNNING\" | \"READY\" | \"FAILED\";\n  /** Diagnostics information to help inspect the states of model or variable. */\n  diagnosticsInfo?: DiagnosticsInfo;\n}\n\n/** An optional field, indicating the manner to align multiple variables. */\nexport interface AlignPolicy {\n  /**\n   * An optional field, indicating how to align different variables to the same\n   * time-range. Either Inner or Outer.\n   */\n  alignMode?: \"Inner\" | \"Outer\";\n  /**\n   * An optional field, indicating how missing values will be filled. One of\n   * Previous, Subsequent, Linear, Zero, Fixed.\n   *\n   * Possible values: Previous, Subsequent, Linear, Zero, Fixed\n   */\n  fillNAMethod?: string;\n  /** An optional field. Required when fillNAMethod is Fixed. */\n  paddingValue?: number;\n}\n\n/** Diagnostics information to help inspect the states of model or variable. */\nexport interface DiagnosticsInfo {\n  /** Model status. */\n  modelState?: ModelState;\n  /** Variable Status. */\n  variableStates?: Array<VariableState>;\n}\n\n/** Model status. */\nexport interface ModelState {\n  /**\n   * This indicates the number of passes of the entire training dataset the\n   * algorithm has completed.\n   */\n  epochIds?: number[];\n  /**\n   * List of metrics used to assess how the model fits the training data for each\n   * epoch.\n   */\n  trainLosses?: number[];\n  /**\n   * List of metrics used to assess how the model fits the validation set for each\n   * epoch.\n   */\n  validationLosses?: number[];\n  /** Latency for each epoch. */\n  latenciesInSeconds?: number[];\n}\n\n/** Request of last detection. */\nexport interface MultivariateLastDetectionOptions {\n  /**\n   * This contains the inference data, including the name, timestamps(ISO 8601) and\n   * values of variables.\n   */\n  variables: Array<VariableValues>;\n  /**\n   * An optional field, which is used to specify the number of top contributed\n   * variables for one anomalous timestamp in the response. The default number is\n   * 10.\n   */\n  topContributorCount: number;\n}\n\n/** Variable values. */\nexport interface VariableValues {\n  /** Variable name of last detection request. */\n  variable: string;\n  /** Timestamps of last detection request */\n  timestamps: string[];\n  /** Values of variables. */\n  values: number[];\n}\n"]}
+{"version":3,"file":"models.js","sourceRoot":"","sources":["../../src/models.ts"],"names":[],"mappings":"AAAA,uCAAuC;AACvC,kCAAkC","sourcesContent":["// Copyright (c) Microsoft Corporation.\n// Licensed under the MIT License.\n\n/** The request of entire or last anomaly detection. */\nexport interface UnivariateDetectionOptions {\n  /**\n   * Time series data points. Points should be sorted by timestamp in ascending\n   * order to match the anomaly detection result. If the data is not sorted\n   * correctly or there is duplicated timestamp, the API will not work. In such\n   * case, an error message will be returned.\n   */\n  series: Array<TimeSeriesPoint>;\n  /**\n   * Optional argument, can be one of yearly, monthly, weekly, daily, hourly,\n   * minutely, secondly, microsecond or none. If granularity is not present, it will\n   * be none by default. If granularity is none, the timestamp property in time\n   * series point can be absent.\n   */\n  granularity?:\n    | \"yearly\"\n    | \"monthly\"\n    | \"weekly\"\n    | \"daily\"\n    | \"hourly\"\n    | \"minutely\"\n    | \"secondly\"\n    | \"microsecond\"\n    | \"none\";\n  /**\n   * Custom Interval is used to set non-standard time interval, for example, if the\n   * series is 5 minutes, request can be set as {\"granularity\":\"minutely\",\n   * \"customInterval\":5}.\n   */\n  customInterval?: number;\n  /**\n   * Optional argument, periodic value of a time series. If the value is null or\n   * does not present, the API will determine the period automatically.\n   */\n  period?: number;\n  /** Optional argument, advanced model parameter, max anomaly ratio in a time series. */\n  maxAnomalyRatio?: number;\n  /**\n   * Optional argument, advanced model parameter, between 0-99, the lower the value\n   * is, the larger the margin value will be which means less anomalies will be\n   * accepted.\n   */\n  sensitivity?: number;\n  /**\n   * Used to specify how to deal with missing values in the input series, it's used\n   * when granularity is not \"none\".\n   *\n   * Possible values: auto, previous, linear, fixed, zero, notFill\n   */\n  imputeMode?: string;\n  /**\n   * Used to specify the value to fill, it's used when granularity is not \"none\"\n   * and imputeMode is \"fixed\".\n   */\n  imputeFixedValue?: number;\n}\n\n/** The definition of input timeseries points. */\nexport interface TimeSeriesPoint {\n  /** Optional argument, timestamp of a data point (ISO8601 format). */\n  timestamp?: Date | string;\n  /** The measurement of that point, should be float. */\n  value: number;\n}\n\n/** The request of change point detection. */\nexport interface UnivariateChangePointDetectionOptions {\n  /**\n   * Time series data points. Points should be sorted by timestamp in ascending\n   * order to match the change point detection result.\n   */\n  series: Array<TimeSeriesPoint>;\n  /**\n   * Can only be one of yearly, monthly, weekly, daily, hourly, minutely or\n   * secondly. Granularity is used for verify whether input series is valid.\n   */\n  granularity:\n    | \"yearly\"\n    | \"monthly\"\n    | \"weekly\"\n    | \"daily\"\n    | \"hourly\"\n    | \"minutely\"\n    | \"secondly\"\n    | \"microsecond\"\n    | \"none\";\n  /**\n   * Custom Interval is used to set non-standard time interval, for example, if the\n   * series is 5 minutes, request can be set as {\"granularity\":\"minutely\",\n   * \"customInterval\":5}.\n   */\n  customInterval?: number;\n  /**\n   * Optional argument, periodic value of a time series. If the value is null or\n   * does not present, the API will determine the period automatically.\n   */\n  period?: number;\n  /**\n   * Optional argument, advanced model parameter, a default stableTrendWindow will\n   * be used in detection.\n   */\n  stableTrendWindow?: number;\n  /**\n   * Optional argument, advanced model parameter, between 0.0-1.0, the lower the\n   * value is, the larger the trend error will be which means less change point will\n   * be accepted.\n   */\n  threshold?: number;\n}\n\n/** ErrorResponse contains code and message that shows the error information. */\nexport interface ErrorResponse {\n  /** The error code. */\n  code: string;\n  /** The message explaining the error reported by the service. */\n  message: string;\n}\n\n/** Variable Status. */\nexport interface VariableState {\n  /** Variable name in variable states. */\n  variable?: string;\n  /** Proportion of missing values that need to be filled by fillNAMethod. */\n  filledNARatio?: number;\n  /** Number of effective data points before applying fillNAMethod. */\n  effectiveCount?: number;\n  /** First valid timestamp with value of input data. */\n  firstTimestamp?: Date | string;\n  /** Last valid timestamp with value of input data. */\n  lastTimestamp?: Date | string;\n}\n\n/**\n * Detection request for batch inference. This is an asynchronous inference which\n * will need another API to get detection results.\n */\nexport interface MultivariateBatchDetectionOptions {\n  /**\n   * Source link to the input data to indicate an accessible Azure storage Uri,\n   * either pointed to an Azure blob storage folder, or pointed to a CSV file in\n   * Azure blob storage based on you data schema selection. The data schema should\n   * be exactly the same with those used in the training phase.\n   */\n  dataSource: string;\n  /**\n   * An optional field, which is used to specify the number of top contributed\n   * variables for one anomalous timestamp in the response. The default number is\n   * 10.\n   */\n  topContributorCount: number;\n  /**\n   * A required field, indicating the start time of data for detection, which should\n   * be date-time of ISO 8601 format.\n   */\n  startTime: Date | string;\n  /**\n   * A required field, indicating the end time of data for detection, which should\n   * be date-time of ISO 8601 format.\n   */\n  endTime: Date | string;\n}\n\n/**\n * Training result of a model including its status, errors and diagnostics\n * information.\n */\nexport interface ModelInfo {\n  /**\n   * Source link to the input data to indicate an accessible Azure storage Uri,\n   * either pointed to an Azure blob storage folder, or pointed to a CSV file in\n   * Azure blob storage based on you data schema selection.\n   */\n  dataSource: string;\n  /**\n   * Data schema of input data source: OneTable or MultiTable. The default\n   * DataSchema is OneTable.\n   *\n   * Possible values: OneTable, MultiTable\n   */\n  dataSchema?: string;\n  /**\n   * A required field, indicating the start time of training data, which should be\n   * date-time of ISO 8601 format.\n   */\n  startTime: Date | string;\n  /**\n   * A required field, indicating the end time of training data, which should be\n   * date-time of ISO 8601 format.\n   */\n  endTime: Date | string;\n  /**\n   * An optional field. The display name of the model whose maximum length is 24\n   * characters.\n   */\n  displayName?: string;\n  /**\n   * An optional field, indicating how many previous timestamps will be used to\n   * detect whether the timestamp is anomaly or not.\n   */\n  slidingWindow?: number;\n  /** An optional field, indicating the manner to align multiple variables. */\n  alignPolicy?: AlignPolicy;\n  /** Model status. One of CREATED, RUNNING, READY, and FAILED. */\n  status?: \"CREATED\" | \"RUNNING\" | \"READY\" | \"FAILED\";\n  /** Diagnostics information to help inspect the states of model or variable. */\n  diagnosticsInfo?: DiagnosticsInfo;\n}\n\n/** An optional field, indicating the manner to align multiple variables. */\nexport interface AlignPolicy {\n  /**\n   * An optional field, indicating how to align different variables to the same\n   * time-range. Either Inner or Outer.\n   */\n  alignMode?: \"Inner\" | \"Outer\";\n  /**\n   * An optional field, indicating how missing values will be filled. One of\n   * Previous, Subsequent, Linear, Zero, Fixed.\n   *\n   * Possible values: Previous, Subsequent, Linear, Zero, Fixed\n   */\n  fillNAMethod?: string;\n  /** An optional field. Required when fillNAMethod is Fixed. */\n  paddingValue?: number;\n}\n\n/** Diagnostics information to help inspect the states of model or variable. */\nexport interface DiagnosticsInfo {\n  /** Model status. */\n  modelState?: ModelState;\n  /** Variable Status. */\n  variableStates?: Array<VariableState>;\n}\n\n/** Model status. */\nexport interface ModelState {\n  /**\n   * This indicates the number of passes of the entire training dataset the\n   * algorithm has completed.\n   */\n  epochIds?: number[];\n  /**\n   * List of metrics used to assess how the model fits the training data for each\n   * epoch.\n   */\n  trainLosses?: number[];\n  /**\n   * List of metrics used to assess how the model fits the validation set for each\n   * epoch.\n   */\n  validationLosses?: number[];\n  /** Latency for each epoch. */\n  latenciesInSeconds?: number[];\n}\n\n/** Request of last detection. */\nexport interface MultivariateLastDetectionOptions {\n  /**\n   * This contains the inference data, including the name, timestamps(ISO 8601) and\n   * values of variables.\n   */\n  variables: Array<VariableValues>;\n  /**\n   * An optional field, which is used to specify the number of top contributed\n   * variables for one anomalous timestamp in the response. The default number is\n   * 10.\n   */\n  topContributorCount: number;\n}\n\n/** Variable values. */\nexport interface VariableValues {\n  /** Variable name of last detection request. */\n  variable: string;\n  /** Timestamps of last detection request */\n  timestamps: string[];\n  /** Values of variables. */\n  values: number[];\n}\n"]}