Predict EdgeCrafter
Overview
Runs an EdgeCrafter model from your workspace registry over an image or video dataset and writes the assets with their predicted boxes or masks to the output dataset. The input dataset stays unchanged.
Pick a Destination dataset to add the predictions to a dataset you already have, such as a labeling pool or an active-learning set: it keeps what it holds, and an image it already annotates gets the new prediction instead. Leave it empty and the node writes into a dataset of its own, one version per run.
The model you pick decides what the node predicts: an ECDet architecture returns bounding boxes, an ECSeg one a mask per instance at the original image resolution as well. Video prediction outputs bounding boxes only. Each prediction carries its score in a confidence tag.
The badge counts the annotated assets and opens the output dataset.
Pretrained Models
- EdgeCrafter ECDet-S
- EdgeCrafter ECDet-M
- EdgeCrafter ECDet-L
- EdgeCrafter ECDet-X
- EdgeCrafter ECSeg-S
- EdgeCrafter ECSeg-M
- EdgeCrafter ECSeg-L
- EdgeCrafter ECSeg-X
Prerequisites
How it works
Takes the destination dataset, or the node’s own output dataset - created on the first run, a new empty version of it on a rerun - and adds the input dataset’s images and videos to it without duplicating them.
FAQ
Does the output dataset keep the input's annotations?
No. The output holds the same assets as the input, but only this run’s predictions as annotations, which is what an evaluation node expects. The input dataset is never changed.
What data can Predict EdgeCrafter process?
Images and videos. If the dataset holds at least one image, only images are predicted: videos are still added to the output dataset but get no predictions in that run. Videos are predicted frame by frame, as boxes even with an ECSeg model.
How do I get masks instead of boxes?
Pick an ECSeg model, either a catalog architecture or a version trained for segmentation. An ECDet one cannot produce masks.
How do I keep only some classes?
Tick the classes to keep in Predict classes, which lists the model’s ontology. Everything else is discarded before the image annotations are written. The filter applies to images only: video predictions keep every class.
What does each prediction record?
Each annotation is tagged with the task that produced it, the model version, and the model’s name. Every prediction stores its score in the numeric confidence tag.
Why is the output empty and the badge 0?
Why is the output empty and the badge 0?
The input dataset has no image or video assets. The output dataset is still created or versioned, but stays empty.
How do I know whether the predictions are any good?
Run Evaluate object detection on box predictions, with the input dataset as ground truth. Prediction alone reports how many assets were annotated, not how correct the annotations are.
Inputs
Dataset to run inference on. Images are annotated when the dataset holds any, videos frame by frame otherwise. Key: DATASET.
Dataset the predictions are added to, keeping what it already holds. An image it already annotates gets the new prediction instead. Leave empty to use the node’s own output dataset, which a rerun writes a new version of. Key: DST_DATASET.
Model to run inference with, from this workspace or the zoo, and with it the task: an ECDet version predicts boxes, an ECSeg one masks as well. The node runs on the resolved version’s weights: the checkpoint stored for it, or the public COCO weights of its architecture for a pretrained entry that hosts none. Key: MODEL.
Classes to keep in the predictions - everything else the model detects is dropped. Leave empty to keep all of them. Values come from MODEL. Key: CLASSES.
Inference settings in YAML - how confident a detection must be to be kept, and how many images go through the GPU at once. Key: CONFIG.
Number of images fetched at a time while the GPU works through the ones already downloaded. Minimum: 1. Key: DOWNLOAD_BATCH_SIZE.
How far downloading may run ahead of inference, counted in batches. Raise it when a slow connection leaves the GPU idle, at the cost of disk space. Minimum: 1. Key: PREFETCH_BATCHES.
Outputs
Dataset with the input images plus the EdgeCrafter prediction annotations: the destination dataset when one is picked, otherwise the node’s own. Shown as an artifact. Key: OUTPUT_DATASET.
Number of images annotated. Shown on the node as a badge. Selecting the badge opens OUTPUT_DATASET. Key: PRED_COUNT.
Models and configuration
Model is the one place the weights come from. The picker lists the workspace’s EdgeCrafter models and their ready versions. Picking the model itself follows its newest ready version at every run, and picking a version pins it. A trained version runs on its stored checkpoint and training config, and a catalog version on the public COCO weights of its architecture, which predict the 80 COCO classes.
The picked architecture also decides the task: an ECDet model predicts boxes, an ECSeg one masks as well. A trained checkpoint carries its own class list, which names the predictions. When it has none, the version’s recorded classes are used.
Config holds the inference settings. conf_threshold (default 0.4) is the score a prediction needs to survive, and batch_size (default 4) how many images or video frames the model sees at once. Config must be a YAML mapping.
Runtime
The node runs in its own GPU container. The first run takes longer while the worker pulls the image, and a catalog model downloads its public weights on every run.
Images are fetched ahead of the model to keep the GPU busy. Download batch size sets how many arrive at once and Prefetch batches how many wait on local disk, so raise them for fast storage and lower them when disk is tight.
JSON config
Machine-readable node interface for automation and advanced usage.