Dataset
dataset-openapi (1.0)
Download OpenAPI specification:Download
Dataset List
Get a list of dataset groups that are created under your team plus all versions under each group.
Responses
Response samples
- 200
[- {
- "id": "string",
- "name": "string",
- "createdAt": "string",
- "versionList": [
- {
- "versionId": "string",
- "createdAt": "string",
- "createdBy": "string",
- "statsSummary": {
- "totalFrames": 0,
- "totalBoxes": 0
}, - "versionNo": "string",
- "versionDetails": "string"
}
]
}
]
Download Dataset
Returns training data and urls of images for all frames (both real and augmented) in given export format
path Parameters
datasetVersionId required | string ID of the required version of relevant dataset group. This id can be taken from response of Dataset List endpoint. |
query Parameters
pageIndex | number Index of the page of data frames (default = 0) |
pageSize | number No of frames contained in each page (default 100 and maximum is 10000) |
exportFormat required | string Output format of data (accepted formats are: yolo, rcnn, tensorflow) |
Create Dataset Version
Creates a dataset version under required dataset group using given projects. You can input the split percentage for 3 categories that are required for ML process - training / testing and validation.
path Parameters
datasetGroupId required | string ID of the dataset group under which this version to be created. This ID can be retrieved from the Dataset List endpoint response. |
Request Body schema: application/json
The percentages to which the dataset should be split for training, testing and validation. Make sure the sum of percentages equals 100.
trainingSetPercentage required | number |
testingSetPercentage required | number |
validationSetPercentage required | number |
Responses
Request samples
- Payload
{- "trainingSetPercentage": 0,
- "testingSetPercentage": 0,
- "validationSetPercentage": 0
}