> ## Documentation Index > Fetch the complete documentation index at: https://docs.edgeimpulse.com/llms.txt > Use this file to discover all available pages before exploring further. # Deploy pretrained model > Takes in a TFLite file and builds the model and SDK. Updates are streamed over the websocket API (or can be retrieved through the /stdout endpoint). Use getProfileTfliteJobResult to get the results when the job is completed. ## OpenAPI ````yaml /.assets/openapi.yaml post /api/{projectId}/jobs/deploy-pretrained-model openapi: 3.0.0 info: title: Edge Impulse API version: 1.0.0 servers: - url: https://studio.edgeimpulse.com/v1 security: - ApiKeyAuthentication: [] - JWTAuthentication: [] - JWTHttpHeaderAuthentication: [] - OAuth2: [] paths: /api/{projectId}/jobs/deploy-pretrained-model: post: tags: - Jobs summary: Deploy pretrained model description: >- Takes in a TFLite file and builds the model and SDK. Updates are streamed over the websocket API (or can be retrieved through the /stdout endpoint). Use getProfileTfliteJobResult to get the results when the job is completed. operationId: startDeployPretrainedModelJob parameters: - $ref: '#/components/parameters/ProjectIdParameter' - $ref: '#/components/parameters/OptionalImpulseIdParameter' requestBody: required: true content: application/json: schema: $ref: '#/components/schemas/DeployPretrainedModelRequest' responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/StartJobResponse' components: parameters: ProjectIdParameter: name: projectId in: path required: true description: Project ID schema: type: integer OptionalImpulseIdParameter: name: impulseId in: query required: false description: Impulse ID. If this is unset then the default impulse is used. schema: type: integer schemas: DeployPretrainedModelRequest: type: object required: - modelFileBase64 - modelFileType - deploymentType - modelInfo properties: modelFileBase64: description: A base64 encoded pretrained model type: string modelFileType: type: string enum: - tflite - onnx - saved_model - lgbm - xgboost - pickle - ngboost deploymentType: type: string description: >- The name of the built target. You can find this by listing all deployment targets through `listDeploymentTargetsForProject` (via `GET /v1/api/{projectId}/deployment/targets`) and see the `format` type. engine: $ref: '#/components/schemas/DeploymentTargetEngine' modelInfo: type: object required: - input - model properties: input: discriminator: propertyName: inputType mapping: time-series: $ref: '#/components/schemas/DeployPretrainedModelInputTimeSeries' audio: $ref: '#/components/schemas/DeployPretrainedModelInputAudio' image: $ref: '#/components/schemas/DeployPretrainedModelInputImage' other: $ref: '#/components/schemas/DeployPretrainedModelInputOther' oneOf: - $ref: '#/components/schemas/DeployPretrainedModelInputTimeSeries' - $ref: '#/components/schemas/DeployPretrainedModelInputAudio' - $ref: '#/components/schemas/DeployPretrainedModelInputImage' - $ref: '#/components/schemas/DeployPretrainedModelInputOther' model: discriminator: propertyName: modelType mapping: classification: $ref: >- #/components/schemas/DeployPretrainedModelModelClassification regression: $ref: '#/components/schemas/DeployPretrainedModelModelRegression' object-detection: $ref: >- #/components/schemas/DeployPretrainedModelModelObjectDetection freeform: $ref: '#/components/schemas/DeployPretrainedModelModelFreeform' anomaly: $ref: '#/components/schemas/DeployPretrainedModelModelAnomaly' visual-anomaly: $ref: >- #/components/schemas/DeployPretrainedModelModelVisualAnomaly oneOf: - $ref: >- #/components/schemas/DeployPretrainedModelModelClassification - $ref: '#/components/schemas/DeployPretrainedModelModelRegression' - $ref: >- #/components/schemas/DeployPretrainedModelModelObjectDetection - $ref: '#/components/schemas/DeployPretrainedModelModelFreeform' - $ref: '#/components/schemas/DeployPretrainedModelModelAnomaly' - $ref: '#/components/schemas/DeployPretrainedModelModelVisualAnomaly' representativeFeaturesBase64: description: >- A base64 encoded .npy file containing the features from your validation set (optional for onnx and saved_model) - used to quantize your model. type: string deployModelType: type: string enum: - int8 - float32 useConverter: description: Optional, use a specific converter (only for ONNX models). type: string enum: - onnx-tf - onnx2tf overrideInputShape: description: >- Optional for ONNX files: overrides the input shape of the model. This is highly suggested if the model has dynamic dimensions. If this field is not set, then all dynamic dimensions will be set to '1'. type: array items: type: integer StartJobResponse: allOf: - $ref: '#/components/schemas/GenericApiResponse' - type: object required: - id properties: id: type: integer description: Job identifier. Status updates will include this identifier. example: 12873488112 DeploymentTargetEngine: type: string enum: - tflite - tflite-eon - tflite-eon-ram-optimized - tensorrt - tensaiflow - drp-ai - tidl - akida - syntiant - memryx - neox - ethos-linux - st-aton - ceva-npn - nordic-axon - vlm-connector DeployPretrainedModelInputTimeSeries: type: object required: - inputType - frequencyHz - windowLengthMs properties: inputType: type: string enum: - time-series frequencyHz: type: number windowLengthMs: type: integer DeployPretrainedModelInputAudio: type: object required: - inputType - frequencyHz properties: inputType: type: string enum: - audio frequencyHz: type: number DeployPretrainedModelInputImage: type: object required: - inputType properties: inputType: type: string enum: - image inputScaling: $ref: '#/components/schemas/ImageInputScaling' resizeMode: $ref: '#/components/schemas/ImageInputResizeMode' DeployPretrainedModelInputOther: type: object required: - inputType properties: inputType: type: string enum: - other DeployPretrainedModelModelClassification: type: object required: - modelType - labels properties: modelType: type: string enum: - classification labels: type: array items: type: string DeployPretrainedModelModelRegression: type: object required: - modelType properties: modelType: type: string enum: - regression DeployPretrainedModelModelObjectDetection: type: object required: - modelType - labels - lastLayer - minimumConfidence properties: modelType: type: string enum: - object-detection labels: type: array items: type: string lastLayer: $ref: '#/components/schemas/ObjectDetectionLastLayer' minimumConfidence: description: >- Deprecated: use thresholdValues instead. Threshold for objects (f.e. 0.3). type: number thresholdValues: type: object description: >- All configured thresholds for the current model. Valid keys are 'min_score' (object detection models, all but paddleocr-detector); 'min_score_pixel', 'min_score_box', 'unclip_ratio' (paddleocr-detector). additionalProperties: type: number DeployPretrainedModelModelFreeform: type: object required: - modelType properties: modelType: type: string enum: - freeform DeployPretrainedModelModelAnomaly: type: object required: - modelType properties: modelType: type: string enum: - anomaly DeployPretrainedModelModelVisualAnomaly: type: object required: - modelType properties: modelType: type: string enum: - visual-anomaly thresholdValues: type: object description: >- All configured thresholds for the current model. Valid keys are 'min_anomaly_score' and 'anomaly_scoring_aggregation_method'. additionalProperties: $ref: '#/components/schemas/ThresholdValue' GenericApiResponse: type: object required: - success properties: success: type: boolean description: Whether the operation succeeded error: type: string description: Optional error description (set if 'success' was false) ImageInputScaling: description: >- Normalization that is applied to images. If this is not set then 0..1 is used. "0..1" gives you non-normalized pixels between 0 and 1. "-1..1" gives you non-normalized pixels between -1 and 1. "0..255" gives you non-normalized pixels between 0 and 255. "-128..127" gives you non-normalized pixels between -128 and 127. "torch" first scales pixels between 0 and 1, then applies normalization using the ImageNet dataset (same as `torchvision.transforms.Normalize()`). "bgr-subtract-imagenet-mean" scales to 0..255, reorders pixels to BGR, and subtracts the ImageNet mean from each channel. type: string enum: - 0..1 - '-1..1' - '-128..127' - 0..255 - torch - bgr-subtract-imagenet-mean ImageInputResizeMode: type: string description: > Input images are resized automatically before training and testing, to match the impulse input shape. This determines the resize mode used when the aspect ratio of the input data is different to the aspect ratio of the impulse. example: squash enum: - squash - fit-short - fit-long - crop ObjectDetectionLastLayer: type: string enum: - mobilenet-ssd - fomo - yolov2-akida - yolov5 - yolov5v5-drpai - yolox - yolov7 - yolo-pro - tao-retinanet - tao-ssd - tao-yolov3 - tao-yolov4 - yolov11 - yolov11-abs - paddleocr-detector - qc-face-det-lite ThresholdValue: description: Current value of the threshold oneOf: - type: number - type: string securitySchemes: ApiKeyAuthentication: type: apiKey in: header name: x-api-key JWTAuthentication: type: apiKey in: cookie name: jwt JWTHttpHeaderAuthentication: type: apiKey in: header name: x-jwt-token OAuth2: type: oauth2 flows: authorizationCode: authorizationUrl: /v1/oauth/authorize tokenUrl: /v1/oauth/token scopes: openid: Access to basic profile information email: Access to email address profile: Access to full profile information implicit: authorizationUrl: /v1/oauth/authorize scopes: openid: Access to basic profile information email: Access to email address profile: Access to full profile information password: tokenUrl: /v1/oauth/token scopes: openid: Access to basic profile information email: Access to email address profile: Access to full profile information clientCredentials: tokenUrl: /v1/oauth/token scopes: openid: Access to basic profile information email: Access to email address profile: Access to full profile information ````