Trex - Embedding Inference

The TRex Embedding Inference algorithm enables user inferring image and get the boxes and scores on the same image by embd files they trained from the Embedding Customization.

This algorithm supports batch inference for multiple images, and for every image, multiple embeddings are supported.

Usage Pattern

First of all, make sure you have installed this SDK by pip:

pip install dds-cloudapi-sdk

Then trigger the customization algorithm by TRexEmbdInfer:

from dds_cloudapi_sdk import Client
from dds_cloudapi_sdk import Config
from dds_cloudapi_sdk import BatchEmbdInfer
from dds_cloudapi_sdk import BatchEmbdPrompt
from dds_cloudapi_sdk import TRexEmbdInfer

token = "You API token here"
config = Config(token)
client = Client(config)

image_url = "https://dev.deepdataspace.com/static/04_a.ae28c1d6.jpg"
embd_url = "https://url/to/your/embd/url.file"  # the embd url trained from TRexEmbdCustomize Task

infer_1 = BatchEmbdInfer(
    image=image_url,
    prompts=[
        BatchEmbdPrompt(embd=embd_url, category_id=1)
    ]
)
task = TRexEmbdInfer([infer_1])
client.run_task(task)

for image_objects in task.result.object_batches:
    for obj in image_objects:
        print(obj.score)
        print(obj.bbox)
        print(obj.category_id)
        break
    break

API Reference

class TRexEmbdInfer(batch_infers)[source]

Trigger the Trex Embedding Inference algorithm.

This task can process prompts from multiple images, and each image can have several embedding prompts.

Parameters:

batch_infers (List[BatchEmbdInfer]) – list of BatchPointInfer objects or BatchRectInfer.

property result: TaskResult

Get the formatted TaskResult object.

class BatchEmbdInfer(*, image, prompts)[source]

An infer image with batch embd prompts.

Parameters:
image: str

the image url to be inferred on

prompts: List[BatchEmbdPrompt]

a list of BatchEmbdPrompt

class BatchEmbdPrompt(*, embd=None, category_id)[source]

A batch of embd prompts.

Parameters:

  • embd (str) – the embedding file url

  • category_id (int) – the category id of the rects

embd: str

the embedding file url

category_id: int

the category id of the objects inferred by this embedding file

class TaskResult(*, object_batches)[source]

The task result of TRexInteractiveInfer task.

Parameters:

object_batches (List[List[TRexObject]]) – a 2D list of detected objects of TRexObject, each inner list is the detected objects of one image

object_batches: List[List[TRexObject]]

a 2D list of detected objects of TRexObject, each inner list is the detected objects of one image

class TRexObject(*, score, bbox, category_id)[source]

The object detected by TRexEmbdInfer task.

Parameters:

  • score (float) – the prediction score

  • bbox (List[float]) – the bounding box, [upper_left_x, upper_left_y, lower_right_x, lower_right_y]

  • category_id (int) – the category id of the object

score: float

the prediction score

bbox: List[float]

the bounding box, [upper_left_x, upper_left_y, lower_right_x, lower_right_y]

category_id: int

the category id of the object