Add Utterance with wit.ai API on New Image Sample Created from Leap API

Pipedream makes it easy to connect APIs for wit.ai, Leap and 2,700+ other apps remarkably fast.

Trigger workflow on

New Image Sample Created from the Leap API

Next, do this

Add Utterance with the wit.ai API

No credit card required

▶

Watch us build a workflow

8 min

Watch now ➜

Trusted by 1,000,000+ developers from startups to Fortune 500 companies

Developers ♥ Pipedream

Getting Started#

This integration creates a workflow with a Leap trigger and wit.ai action. When you configure and deploy the workflow, it will run on Pipedream's servers 24x7 for free.

Select this integration
Configure the New Image Sample Created trigger
1. Connect your Leap account
2. Configure timer
3. Select a Model
Configure the Add Utterance action
1. Connect your wit.ai account
2. Configure Text
3. Optional- Configure Out of Scope
Deploy the workflow
Send a test event to validate your setup
Turn on the trigger

Details#

This integration uses pre-built, source-available components from Pipedream's GitHub repo. These components are developed by Pipedream and the community, and verified and maintained by Pipedream.

To contribute an update to an existing component or create a new component, create a PR on GitHub. If you're new to Pipedream component development, you can start with quickstarts for trigger span and action development, and then review the component API reference.

Trigger#

New Image Sample Created on Leap

Description:Emit new event when a new image sample is created for a model. [See the documentation](https://docs.tryleap.ai/reference/samplescontroller_findall-1)

Version:0.0.1

Key:leap-new-image-sample-created

View on GitHub

Leap Overview#

The Leap API enables automated interactions with the Leap.ai platform, which focuses on matching users with optimal job opportunities based on skills and preferences. In Pipedream, you can harness this API to create workflows that streamline the job search process, manage and analyze job matching data, or even integrate with other platforms to enhance the job seeking experience. With Pipedream's serverless execution environment, you can trigger these workflows on a schedule, via webhooks, or in response to events from other apps.

Trigger Code#

import common from "../common/base.mjs";

export default {
  ...common,
  key: "leap-new-image-sample-created",
  name: "New Image Sample Created",
  description: "Emit new event when a new image sample is created for a model. [See the documentation](https://docs.tryleap.ai/reference/samplescontroller_findall-1)",
  version: "0.0.1",
  type: "source",
  dedupe: "unique",
  props: {
    ...common.props,
    model: {
      propDefinition: [
        common.props.leap,
        "model",
      ],
    },
  },
  methods: {
    ...common.methods,
    async processHistoricalEvents({ limit }) {
      const images = await this.leap.listImages({
        modelId: this.model,
      });
      if (!images?.length) {
        return;
      }

      let maxLastCreated = 0;
      for (const image of images.slice(limit * -1)) {
        this.emitEvent(image);
        const ts = Date.parse(image.createdAt);
        if (ts > maxLastCreated) {
          maxLastCreated = ts;
        }
      }
      this._setPrevious(maxLastCreated);
    },
    generateMeta(image) {
      return {
        id: image.id,
        summary: image.uri,
        ts: Date.parse(image.createdAt),
      };
    },
    async processEvent() {
      const lastCreated = this._getPrevious() || 0;
      let maxLastCreated = lastCreated;

      const images = await this.leap.listImages({
        modelId: this.model,
      });
      if (!images.length) {
        return;
      }
      for (const image of images) {
        const ts = Date.parse(image.createdAt);
        if (ts > lastCreated) {
          this.emitEvent(image);
          if (ts > maxLastCreated) {
            maxLastCreated = ts;
          }
        }
      }
      this._setPrevious(maxLastCreated);
    },
  },
};

Trigger Configuration#

This component may be configured based on the props defined in the component code. Pipedream automatically prompts for input values in the UI and CLI.

Label	Prop	Type	Description
Leap	`leap`	`app`	This component uses the Leap app.
N/A	`db`	`$.service.db`	This component uses `$.service.db` to maintain state between executions.
	`timer`	`$.interface.timer`
Model	`model`	`string`	Select a value from the drop down menu.

Trigger Authentication#

Leap uses API keys for authentication. When you connect your Leap account, Pipedream securely stores the keys so you can easily authenticate to Leap APIs in both code and no-code steps.

To retrieve your API Keys,

Navigate to your Leap account and sign in
Click on one of your Projects on the top left dropdown
Go to “API Keys”

About Leap#

Add AI features to your app in minutes. Generate images, fine tune models, and more with our easy-to-use API.

Action#

Add Utterance on wit.ai

Description:Add a sample utterance to train your app with an intent. [See the documentation](https://wit.ai/docs/http/20240304#post__utterances).

Version:0.0.1

Key:wit_ai-add-utterance

View on GitHub

wit.ai Overview#

Wit.ai API enables your apps to process natural language and understand human commands contextually. With it, you can build conversational interfaces, decipher intent from text or voice inputs, and train your model to improve recognition. Pipedream provides a seamless platform to trigger workflows using this API, allowing the automation of tasks based on language cues. It's perfect for crafting custom bots, enhancing customer service, or streamlining internal operations by interpreting human requests and taking action accordingly.

Action Code#

import app from "../../wit_ai.app.mjs";
import utils from "../../common/utils.mjs";
import constants from "../../common/constants.mjs";

export default {
  key: "wit_ai-add-utterance",
  name: "Add Utterance",
  description: "Add a sample utterance to train your app with an intent. [See the documentation](https://wit.ai/docs/http/20240304#post__utterances).",
  version: "0.0.1",
  type: "action",
  props: {
    app,
    text: {
      type: "string",
      label: "Text",
      description: "The text (sentence) you want your app to understand.",
    },
    outOfScope: {
      type: "boolean",
      label: "Out of Scope",
      description: "To train your app to recognize utterances that it should not handle set this to `true`.",
      optional: true,
      default: false,
      reloadProps: true,
    },
  },
  async additionalProps() {
    if (this.outOfScope === true) {
      return {};
    }

    const data = await this.app.listIntents();
    const intents = data.map(({ name }) => name);
    const intentOptions = intents.concat(Object.values(constants.BUILTIN_INTENTS));

    return {
      intent: {
        type: "string",
        label: "Intent",
        description: "The name of the intent you wish to create or train.",
        options: intentOptions,
      },
      entities: {
        type: "string[]",
        label: "Entities",
        description: "The list of entities appearing in this sentence, that you want your app to extract once it is trained. Each entity must be a JSON string with the following properties:\n- `entity` (string, required): The entity name, including the role (e.g., `temperature:temperature` or `wit$temperature:temperature`).\n- `start` (integer, required): The starting index within the text.\n- `end` (integer, required): The ending index within the text.\n- `body` (string, required): The span of the text for the entity.\n- `entities` (array, required): List of entities found within the composite entity.\n\nExample: `{\"entity\":\"wit$temperature:temperature\",\"start\":19,\"end\":26,\"body\":\"34 degrees\",\"entities\":[]}`",
        optional: true,
      },
      traits: {
        type: "string[]",
        label: "Traits",
        description: "The list of traits that are relevant for the utterance. Each trait must be a JSON string with the following properties:\n- `trait` (string, required): The trait name. This can be a trait you created, or a built-in one. i.e `faq` or `wit$sentiment`.\n- `value` (string, required): The value for the trait, e.g. `positive`.\n\nExample: `{\"trait\":\"wit$sentiment\",\"value\":\"neutral\"}`",
        optional: true,
      },
    };
  },
  methods: {
    addUtterance(args = {}) {
      return this.app.post({
        path: "/utterances",
        ...args,
      });
    },
  },
  async run({ $ }) {
    const {
      addUtterance,
      text,
      intent,
      entities,
      traits,
    } = this;

    const response = await addUtterance({
      $,
      data: [
        {
          text,
          intent,
          entities: utils.parseArray(entities),
          traits: utils.parseArray(traits),
        },
      ],
    });

    $.export("$summary", "Successfully added an utterance to your app.");
    return response;
  },
};

Action Configuration#

This component may be configured based on the props defined in the component code. Pipedream automatically prompts for input values in the UI.

Label	Prop	Type	Description
wit.ai	`app`	`app`	This component uses the wit.ai app.
Text	`text`	`string`	The text (sentence) you want your app to understand.
Out of Scope	`outOfScope`	`boolean`	To train your app to recognize utterances that it should not handle set this to `true`.