Detect room features from images (hardwood floor, granite countertop, kitchen island, etc.)
Can process up to 30 image urls at a time.

Detect Features

Detect room function from images (living room, kitchen, etc.)
Can process up to 30 image urls at a time.

Detect Room Function

Generate alt text from images. Uses detected features and room function.
Can process up to 30 image urls at a time.

Generate Alt Text

Score interior images in terms of quality and condition.
Ranges is from 0 (disrepair) to 1 (excellent).
Can process up to 30 image urls at a time.

Score Interior

Computer Vision

List property report subscriptions that belong to the current user, or all reports for the organization.

Set this to true to get all reports that belong to the organization, and not just the ones for the current user

List

Property Report

The query string. Represents the property name, address, etc. We'll also surface aliases and other names, as well as other known addresses.

The expected state. When specified, matched properties will be filtered by state. For example, 'CA' for California.

The zip code. When specified, matched properties will be filtered by zip code. Accept both 5 and ZIP+4 digits formats (e.g. 12345 or 12345-6789). If a leading zero is missing, it will be added automatically.

Search

The property id (UUIDv5 retrieved from the search endpoint).

Get Property

The tier of the property. Either 'valueAdd' or 'current'. When 'valueAdd', the results will assume a better condition of the property.

The renovations to apply to the property. A value between 0 and 1 to add to the property quality. (e.g. 0.1 will transform a 0.4 quality property into a 0.5 quality property).

The maximum distance in miles between the subject and the rent comps.

The minimum distance in miles between the subject and the rent comps.

The minimum number of units of the rent comps. Does not filter properties with no units.

The maximum number of units of the rent comps. Does not filter properties with no units.

Whether to allow properties with no units to be included in the results. False by default.

The minimum year built of the rent comps. Does not filter properties with no year built.

The maximum year built of the rent comps. Does not filter properties with no year built.

Whether to allow properties with no year built to be included in the results. False by default.

The minimum number of stories of the rent comps. Does not filter properties with no stories.

The maximum number of stories of the rent comps. Does not filter properties with no stories.

Whether to allow properties with no stories to be included in the results. False by default.

Comparables

Insert a new property in the database. For example a new development.

Post Property

Get pricing recommendations for active units among a subject property given its details. We use all property and unit details (characteristics, amenities, location, etc.)
to find similar units in the market that have been rented recently. We then use the rent prices of these units to estimate the unit pricing.

Pricing

Get the expense benchmarks for a property given its details.
Our models use all property details (characteristics, #stories, amenities, location, etc.) to compute the estimates.

Expense Benchmarks

Property

Market Search

Market

AccessUser

AltTextResponse

A list of image URLs that were processed.

A list of alt text for each image.
The list is in the same order as the input image URLs.

Describes the status and details of a unit or floorplan's availability,
alongside various attributes like pricing and size. Contains historical availabilty and pricing data.
In some rare cases, the availability object may represent a floorplan rather than a specific unit, when the listing advertises a generic floorplan.

Availability

Flag indicating if the listing is for a generic floorplan rather than a specific unit.
This happens when the listing advertises a generic floorplan rather than a specific unit.

The name of the floorplan, if this entry represents a floorplan. Or the unit's floorplan name.

The specific name or number of the unit, if applicable.

The date the unit was first listed for rent. If null, the unit existed on 06-17-2023, when we first started collecting data, but we cannot say for certain on what day it first appeared.

The last date the unit was listed for rent. If null, the unit is still active.

An array of tags that were predicted using floorplan images, floorplan names and unit amenities

The list of prices for the unit/floorplan over time.

An array detailing the periods during which this unit was active. For price changes within each period, refer to the 'history' key.

The floor number for the unit, derived from the patterns in the unit names among the building.

Date when the unit will become or became available.

The lowest advertised rent for the unit or floorplan.

The asking rent for the unit or the lowest price in a range. Represents the latest asking rent we have seen for the unit.
If you are looking for the asking rent over time, refer to the 'history' key.

The highest advertised rent for the unit or floorplan.

The lowest effective rent after considering discounts or promotions.

The effective rent after considering discounts or promotions. Represents the latest effective price we have seen for the unit.
If you are looking for the effective price over time, refer to the 'history' key.

The highest effective rent after considering discounts or promotions.

The minimum square footage of the unit or floorplan.

The maximum square footage of the unit or floorplan.

Indicates whether the unit has been renovated, extracted from the floorplan name.

Indicates whether the unit is furnished, extracted from the floorplan name.

Indicates whether the unit includes a den, extracted from the floorplan name.

Indicates whether the unit is convertible, extracted from the floorplan name.

The date when this record was last updated.

The list of latest lease terms for the unit/floorplan

An array of non-normalized amenities strings direct from property websites

The latest security deposit for this unit.

AvailabilityPeriod

The date the unit was first listed for rent. If null, the unit existed on 06-17-2023, when we first started collecting data but we cannot say for certain on what day it first appeared.

The list of latest lease terms for the unit/floorplan's availability period

The security deposit of this unit during this period.

The AvgUnitMix object is used to communicate an average unit mix for a
property or market. It includes the average number of beds, baths, and
square footage, as well as the average rent and frequency of occurrence.

AvgUnitMix

The average number of beds for the units in the mix.

The average number of baths for the units in the mix.

The average square footage for the units in the mix.

The average market rent for the units in the mix.

The frequency of occurrence of the unit mix in the dataset.

The frequency of occurrence of the unit mix in the population.

BuildingAmenity

Defines the quality scores for various aspects of a property, including both general areas and specific features.
Scores are based on the assessment of finishes, conditions, and amenities, derived from interior and exterior photo evaluations.
Each score ranges from 0 to 1, where 0 indicates the lowest quality and 1 the highest.

BuildingQuality

The overall quality score of the property, reflecting an aggregate assessment of the entire property.

The average quality score for all bedrooms, assessing finishes, condition, and amenities within these spaces.

The average quality score for the kitchen, evaluating the finishes, appliances, and layout.

The average quality score for all bathrooms, focusing on finishes, fixtures, and condition.

The average quality score for the dining room, considering the space, finishes, and lighting.

The average quality score for common areas, including lobbies, gardens, and other shared spaces.

The average quality score for the fitness center, assessing equipment quality, space, and maintenance.

The average quality score for the laundry room, focusing on the quality of machines and the maintenance of the space.

The average quality score for the living room, evaluating the comfort, finishes, and layout.

The average quality score for the main entrance, considering its appearance, security, and accessibility.

The average quality score for stairs and hallways, assessing the condition, lighting, and safety features.

The average quality score for the swimming pool, if applicable, focusing on maintenance, size, and amenities.

The average quality score for walk-in closets, evaluating the size, organization options, and finishes.

BETA: the object is currently in beta and may change without notice.
The CompetitionProfile object is used to explain the competition profile of a property.
It describes how many properties are in the high, medium, and low similarity buckets within a given radius (non overlapping, non-cumulative).

CompetitionProfile

The radius, in miles, used to compute the competition profile.

The number of properties in the high similarity bucket. (similarity in [0.8, 1])

The number of properties in the medium similarity bucket. (similarity in [0.6, 0.8[)

The number of properties in the low similarity bucket. (similarity in [0, 0.6[)

CompsResponse

Defines optional settings for computing comps. These settings control the scope and criteria of
the comparison. We advise to leave these settings empty to get the most accurate results, but
they can be used to enforce specific criteria for the comparison.

CompsSettings

The maximum search radius (in miles) for comparable properties. This limits how far out
from the subject property the search for comparables will go.

Force the search to only consider units having a specific status (active, closed, or all).
By default, all units are considered.

The start date for the period within which comparable units should be considered.
This filters the data to include only those units occurring after this date.

The end date for the period within which comparable units should be considered.
This filters the data to include only units occurring before this date.

Determines whether to include outlier data in the comparison. Outliers can significantly
affect the average and median values of the comps.

A list of building IDs that should be explicitly included in the comps analysis,
regardless of other criteria.

A list of unit IDs to be excluded from the comps analysis. This can help remove
known anomalies or non-comparable units from the dataset.

Indicates whether the subject property itself should be excluded from the comps analysis.
This is useful when the subject property's data might skew the comparison.

Concessions

The start date from which the special offer was being marketed.

The end date when the special was no longer being marketed.

One concession can contain a lot of information. We break it out depending on condition qualifiers.

ConcessionsItem

Deadline condition for the concession, formatted as a string.

The deadline until which free months are offered, formatted as a string.

Indicates if the application fee is waived.

Indicates if the security deposit is waived.

Indicates if the administrative fee is waived.

Indicates if a cheaper security deposit is available.

Indicates if a cheaper administrative fee is available.

Indicates if a cheaper application fee is available.

Indicates if a cheaper move-in fee is available.

Indicates if the concession is specific to certain units but doesn't say which.

ControllerResponse40UserAttributesAndReportIdsNumberArray41Array

ControllerResponse40UserAttributesAndReportIdsNumberArray41ArrayDataItem

ControllerResponseAccessUserArray

ControllerResponseBoolean

ControllerResponseMarketReportSubscriptionAttributesArray

ControllerResponseMarketReportSubscriptionAttributesOrNull

ControllerResponseNoteAttributesArray

ControllerResponseNoteAttributesOrNull

ControllerResponseNull

ControllerResponseNumberOrNull

ControllerResponseOrganizationAttributesOrNull

ControllerResponsePerSeatBooleanTrialEndStringOrNullCancelAtStringOrNullPastDueBooleanProductsStringArrayArray

ControllerResponsePerSeatBooleanTrialEndStringOrNullCancelAtStringOrNullPastDueBooleanProductsStringArrayArrayDataItem

ControllerResponseRentSourceBulkDataSubscriptionAttributesArray

ControllerResponseRentSourceBulkDataSubscriptionAttributesOrNull

ControllerResponseRentSourceSubscriptionAttributesArray

ControllerResponseRentSourceSubscriptionAttributesOrNull

ControllerResponseString

ControllerResponseUserAttributesArray

DateStr

The Distribution object represents a predicted value.
It represents the lower, predicted, and upper bounds of the predicted value,
as well as the impact of each feature on the predicted value.

Distribution

The lower bound of the predicted value.
Usually represents the 20th percentile of the distribution.

The upper bound of the predicted value.
Usually represents the 80th percentile of the distribution.

An explanation of the impact of each feature on the predicted value.
The impact is a number representing the change in the predicted value for a one-unit change in the feature.
The value is the actual value of the feature.

DistributionExplanationValue

The ExpenseBenchmarksResponse object contains the predicted expense benchmarks for a property.

ExpenseBenchmarksResponse

ExpenseBenchmarksResponseSubjectEstimate

FeatureFlag

A list of detected features and their probabilities.

FeatureResponse

A list of detected features and their probabilities for each image.
The list is in the same order as the input image URLs.

FeatureResponseFeaturesItemItem

The probability that the feature is present in the image.
A number between 0 and 1.

ForecastPayload

ForecastPayloadTask

ForecastResponse

GetPropertyRequestNotFoundErrorBody

The request body for a computer vision request.
A list of up to 30 image URLs to send to the computer vision API.

ImageUrls

A list of image URLs to send to the computer vision API.

InteriorScoreResponse

The score of the property's interior. It's a weighted average of the room qualities by their individual weights, calibrated based on observed correlations with rents.
Feel free to build your own weights based on your business logic, for example if you want to give more importance to the kitchen quality than the bathroom quality.
A number between 0 and 1.

A list of detected qualities and their probabilities for each image.
The list is in the same order as the input image URLs.

InteriorScoreResponseQualitiesItem

A list of detected features and their probabilities for the image.

The quality of the room.
A number between 0 and 1.

InteriorScoreResponseQualitiesItemFeaturesItemItem

LiquidRentSubscription

The ListingStatus enum represents the possible statuses of a listing.

ListingStatus

MarketComparablePayload contains the data needed to find similar markets to a given market.
If the coordinates are not provided, the embeddings are used to find similar markets across the entire dataset.
If the coordinates are provided, the embeddings are used to find similar markets within the specified area.

MarketComparablePayload

The coordinates of the area to find similar markets within. Can be any polygon shape.

The embeddings of the market to find similar markets to.

A list of 2-letter state codes to filter the search to only include markets in those states.

The MarketComparableResponse object is used to retrieve similar markets to a given market.

MarketComparableResponse

The similarity score between the subject and the comparable market. The score ranges from 0 to 1.

The centroid of the census tract. The first element is the latitude, and the second element is the longitude.

The geometry of the census tract. The type is always "Polygon".

Some valuable demographic information about the census tract.

MarketComparableResponseCentroid

MarketComparableResponseGeometry

Interface representing the filters that can be applied to a market search.

1	curl -X POST https://api.hellodata.ai/qualityscore/detect_room_function \
2	-H "x-api-key: <apiKey>" \
3	-H "Content-Type: application/json" \
4	-d '{
5	"image_urls": [
6	"https://example.com/image1.jpg",
7	"https://example.com/image2.jpg"
8	]
9	}'

1	{
2	"result": {
3	"images": [
4	"images"
5	],
6	"rooms": [
7	[
8	{
9	"probability": 0.9,
10	"room": "kitchen"
11	}
12	]
13	]
14	}
15	}