inference-models 0.18.3__py3-none-any.whl

inference_models/models/vit/vit_classification_trt.py

@@ -0,0 +1,365 @@
+from threading import Lock
+from typing import List, Optional, Union
+
+import numpy as np
+import torch
+
+from inference_models import (
+    ClassificationModel,
+    ClassificationPrediction,
+    MultiLabelClassificationModel,
+    MultiLabelClassificationPrediction,
+)
+from inference_models.configuration import DEFAULT_DEVICE
+from inference_models.entities import ColorFormat
+from inference_models.errors import (
+    CorruptedModelPackageError,
+    MissingDependencyError,
+    ModelRuntimeError,
+)
+from inference_models.models.base.types import PreprocessedInputs
+from inference_models.models.common.cuda import (
+    use_cuda_context,
+    use_primary_cuda_context,
+)
+from inference_models.models.common.model_packages import get_model_package_contents
+from inference_models.models.common.roboflow.model_packages import (
+    InferenceConfig,
+    ResizeMode,
+    TRTConfig,
+    parse_class_names_file,
+    parse_inference_config,
+    parse_trt_config,
+)
+from inference_models.models.common.roboflow.pre_processing import (
+    pre_process_network_input,
+)
+from inference_models.models.common.trt import (
+    get_engine_inputs_and_outputs,
+    infer_from_trt_engine,
+    load_model,
+)
+
+try:
+    import tensorrt as trt
+except ImportError as import_error:
+    raise MissingDependencyError(
+        message=f"Could not import YOLOv8 model with TRT backend - this error means that some additional dependencies "
+        f"are not installed in the environment. If you run the `inference-models` library directly in your Python "
+        f"program, make sure the following extras of the package are installed: `trt10` - installation can only "
+        f"succeed for Linux and Windows machines with Cuda 12 installed. Jetson devices, should have TRT 10.x "
+        f"installed for all builds with Jetpack 6. "
+        f"If you see this error using Roboflow infrastructure, make sure the service you use does support the model. "
+        f"You can also contact Roboflow to get support.",
+        help_url="https://todo",
+    ) from import_error
+
+try:
+    import pycuda.driver as cuda
+except ImportError as import_error:
+    raise MissingDependencyError(
+        message="TODO", help_url="https://todo"
+    ) from import_error
+
+
+class VITForClassificationTRT(ClassificationModel[torch.Tensor, torch.Tensor]):
+
+    @classmethod
+    def from_pretrained(
+        cls,
+        model_name_or_path: str,
+        device: torch.device = DEFAULT_DEVICE,
+        engine_host_code_allowed: bool = False,
+        **kwargs,
+    ) -> "VITForClassificationTRT":
+        if device.type != "cuda":
+            raise ModelRuntimeError(
+                message=f"TRT engine only runs on CUDA device - {device} device detected.",
+                help_url="https://todo",
+            )
+        model_package_content = get_model_package_contents(
+            model_package_dir=model_name_or_path,
+            elements=[
+                "class_names.txt",
+                "inference_config.json",
+                "trt_config.json",
+                "engine.plan",
+            ],
+        )
+        class_names = parse_class_names_file(
+            class_names_path=model_package_content["class_names.txt"]
+        )
+        inference_config = parse_inference_config(
+            config_path=model_package_content["inference_config.json"],
+            allowed_resize_modes={
+                ResizeMode.STRETCH_TO,
+                ResizeMode.LETTERBOX,
+                ResizeMode.CENTER_CROP,
+                ResizeMode.LETTERBOX_REFLECT_EDGES,
+            },
+        )
+        if inference_config.post_processing.type != "softmax":
+            raise CorruptedModelPackageError(
+                message="Expected Softmax to be the post-processing",
+                help_url="https://todo",
+            )
+        trt_config = parse_trt_config(
+            config_path=model_package_content["trt_config.json"]
+        )
+        cuda.init()
+        cuda_device = cuda.Device(device.index or 0)
+        with use_primary_cuda_context(cuda_device=cuda_device) as cuda_context:
+            engine = load_model(
+                model_path=model_package_content["engine.plan"],
+                engine_host_code_allowed=engine_host_code_allowed,
+            )
+            execution_context = engine.create_execution_context()
+        inputs, outputs = get_engine_inputs_and_outputs(engine=engine)
+        if len(inputs) != 1:
+            raise CorruptedModelPackageError(
+                message=f"Implementation assume single model input, found: {len(inputs)}.",
+                help_url="https://todo",
+            )
+        if len(outputs) != 1:
+            raise CorruptedModelPackageError(
+                message=f"Implementation assume single model output, found: {len(outputs)}.",
+                help_url="https://todo",
+            )
+        return cls(
+            engine=engine,
+            input_name=inputs[0],
+            output_name=outputs[0],
+            class_names=class_names,
+            inference_config=inference_config,
+            trt_config=trt_config,
+            device=device,
+            cuda_context=cuda_context,
+            execution_context=execution_context,
+        )
+
+    def __init__(
+        self,
+        engine: trt.ICudaEngine,
+        input_name: str,
+        output_name: str,
+        class_names: List[str],
+        inference_config: InferenceConfig,
+        trt_config: TRTConfig,
+        device: torch.device,
+        cuda_context: cuda.Context,
+        execution_context: trt.IExecutionContext,
+    ):
+        self._engine = engine
+        self._input_name = input_name
+        self._output_names = [output_name]
+        self._class_names = class_names
+        self._inference_config = inference_config
+        self._trt_config = trt_config
+        self._device = device
+        self._cuda_context = cuda_context
+        self._execution_context = execution_context
+        self._lock = Lock()
+
+    @property
+    def class_names(self) -> List[str]:
+        return self._class_names
+
+    def pre_process(
+        self,
+        images: Union[torch.Tensor, List[torch.Tensor], np.ndarray, List[np.ndarray]],
+        input_color_format: Optional[ColorFormat] = None,
+        **kwargs,
+    ) -> torch.Tensor:
+        return pre_process_network_input(
+            images=images,
+            image_pre_processing=self._inference_config.image_pre_processing,
+            network_input=self._inference_config.network_input,
+            target_device=self._device,
+            input_color_format=input_color_format,
+        )[0]
+
+    def forward(
+        self, pre_processed_images: PreprocessedInputs, **kwargs
+    ) -> torch.Tensor:
+        with self._lock:
+            with use_cuda_context(context=self._cuda_context):
+                return infer_from_trt_engine(
+                    pre_processed_images=pre_processed_images,
+                    trt_config=self._trt_config,
+                    engine=self._engine,
+                    context=self._execution_context,
+                    device=self._device,
+                    input_name=self._input_name,
+                    outputs=self._output_names,
+                )[0]
+
+    def post_process(
+        self,
+        model_results: torch.Tensor,
+        **kwargs,
+    ) -> ClassificationPrediction:
+        if self._inference_config.post_processing.fused:
+            confidence = model_results
+        else:
+            confidence = torch.nn.functional.softmax(model_results, dim=-1)
+        return ClassificationPrediction(
+            class_id=confidence.argmax(dim=-1),
+            confidence=confidence,
+        )
+
+
+class VITForMultiLabelClassificationTRT(
+    MultiLabelClassificationModel[torch.Tensor, torch.Tensor]
+):
+
+    @classmethod
+    def from_pretrained(
+        cls,
+        model_name_or_path: str,
+        device: torch.device = DEFAULT_DEVICE,
+        engine_host_code_allowed: bool = False,
+        **kwargs,
+    ) -> "VITForMultiLabelClassificationTRT":
+        if device.type != "cuda":
+            raise ModelRuntimeError(
+                message=f"TRT engine only runs on CUDA device - {device} device detected.",
+                help_url="https://todo",
+            )
+        model_package_content = get_model_package_contents(
+            model_package_dir=model_name_or_path,
+            elements=[
+                "class_names.txt",
+                "inference_config.json",
+                "trt_config.json",
+                "engine.plan",
+            ],
+        )
+        class_names = parse_class_names_file(
+            class_names_path=model_package_content["class_names.txt"]
+        )
+        inference_config = parse_inference_config(
+            config_path=model_package_content["inference_config.json"],
+            allowed_resize_modes={
+                ResizeMode.STRETCH_TO,
+                ResizeMode.LETTERBOX,
+                ResizeMode.CENTER_CROP,
+                ResizeMode.LETTERBOX_REFLECT_EDGES,
+            },
+        )
+        if inference_config.post_processing.type != "sigmoid":
+            raise CorruptedModelPackageError(
+                message="Expected sigmoid to be the post-processing",
+                help_url="https://todo",
+            )
+        trt_config = parse_trt_config(
+            config_path=model_package_content["trt_config.json"]
+        )
+        cuda.init()
+        cuda_device = cuda.Device(device.index or 0)
+        with use_primary_cuda_context(cuda_device=cuda_device) as cuda_context:
+            engine = load_model(
+                model_path=model_package_content["engine.plan"],
+                engine_host_code_allowed=engine_host_code_allowed,
+            )
+            execution_context = engine.create_execution_context()
+        inputs, outputs = get_engine_inputs_and_outputs(engine=engine)
+        if len(inputs) != 1:
+            raise CorruptedModelPackageError(
+                message=f"Implementation assume single model input, found: {len(inputs)}.",
+                help_url="https://todo",
+            )
+        if len(outputs) != 1:
+            raise CorruptedModelPackageError(
+                message=f"Implementation assume single model output, found: {len(outputs)}.",
+                help_url="https://todo",
+            )
+        return cls(
+            engine=engine,
+            input_name=inputs[0],
+            output_name=outputs[0],
+            class_names=class_names,
+            inference_config=inference_config,
+            trt_config=trt_config,
+            device=device,
+            cuda_context=cuda_context,
+            execution_context=execution_context,
+        )
+
+    def __init__(
+        self,
+        engine: trt.ICudaEngine,
+        input_name: str,
+        output_name: str,
+        class_names: List[str],
+        inference_config: InferenceConfig,
+        trt_config: TRTConfig,
+        device: torch.device,
+        cuda_context: cuda.Context,
+        execution_context: trt.IExecutionContext,
+    ):
+        self._engine = engine
+        self._input_name = input_name
+        self._output_names = [output_name]
+        self._class_names = class_names
+        self._inference_config = inference_config
+        self._trt_config = trt_config
+        self._device = device
+        self._cuda_context = cuda_context
+        self._execution_context = execution_context
+        self._lock = Lock()
+
+    @property
+    def class_names(self) -> List[str]:
+        return self._class_names
+
+    def pre_process(
+        self,
+        images: Union[torch.Tensor, List[torch.Tensor], np.ndarray, List[np.ndarray]],
+        input_color_format: Optional[ColorFormat] = None,
+        **kwargs,
+    ) -> torch.Tensor:
+        return pre_process_network_input(
+            images=images,
+            image_pre_processing=self._inference_config.image_pre_processing,
+            network_input=self._inference_config.network_input,
+            target_device=self._device,
+            input_color_format=input_color_format,
+        )[0]
+
+    def forward(
+        self, pre_processed_images: PreprocessedInputs, **kwargs
+    ) -> torch.Tensor:
+        with self._lock:
+            with use_cuda_context(context=self._cuda_context):
+                return infer_from_trt_engine(
+                    pre_processed_images=pre_processed_images,
+                    trt_config=self._trt_config,
+                    engine=self._engine,
+                    context=self._execution_context,
+                    device=self._device,
+                    input_name=self._input_name,
+                    outputs=self._output_names,
+                )[0]
+
+    def post_process(
+        self,
+        model_results: torch.Tensor,
+        confidence: float = 0.5,
+        **kwargs,
+    ) -> List[MultiLabelClassificationPrediction]:
+        if self._inference_config.post_processing.fused:
+            model_results = model_results
+        else:
+            model_results = torch.nn.functional.sigmoid(model_results)
+        results = []
+        for batch_element_confidence in model_results:
+            predicted_classes = torch.argwhere(
+                batch_element_confidence >= confidence
+            ).squeeze(dim=-1)
+            results.append(
+                MultiLabelClassificationPrediction(
+                    class_ids=predicted_classes,
+                    confidence=batch_element_confidence,
+                )
+            )
+        return results

inference_models/models/yolact/__init__.py

@@ -0,0 +1 @@
+# TODO: decide if port is needed

inference_models/models/yolact/yolact_instance_segmentation_onnx.py

@@ -0,0 +1,336 @@
+from threading import Lock
+from typing import List, Optional, Tuple, Union
+
+import numpy as np
+import torch
+import torchvision
+
+from inference_models import InstanceDetections, InstanceSegmentationModel
+from inference_models.configuration import DEFAULT_DEVICE
+from inference_models.entities import ColorFormat
+from inference_models.errors import (
+    EnvironmentConfigurationError,
+    MissingDependencyError,
+    ModelRuntimeError,
+)
+from inference_models.models.common.model_packages import get_model_package_contents
+from inference_models.models.common.onnx import (
+    run_session_with_batch_size_limit,
+    set_execution_provider_defaults,
+)
+from inference_models.models.common.roboflow.model_packages import (
+    InferenceConfig,
+    PreProcessingMetadata,
+    ResizeMode,
+    parse_class_names_file,
+    parse_inference_config,
+)
+from inference_models.models.common.roboflow.post_processing import (
+    align_instance_segmentation_results,
+    crop_masks_to_boxes,
+)
+from inference_models.models.common.roboflow.pre_processing import (
+    pre_process_network_input,
+)
+from inference_models.utils.onnx_introspection import (
+    get_selected_onnx_execution_providers,
+)
+
+try:
+    import onnxruntime
+except ImportError as import_error:
+    raise MissingDependencyError(
+        message=f"Could not import YOLOv5 model with ONNX backend - this error means that some additional dependencies "
+        f"are not installed in the environment. If you run the `inference-models` library directly in your Python "
+        f"program, make sure the following extras of the package are installed: \n"
+        f"\t* `onnx-cpu` - when you wish to use library with CPU support only\n"
+        f"\t* `onnx-cu12` - for running on GPU with Cuda 12 installed\n"
+        f"\t* `onnx-cu118` - for running on GPU with Cuda 11.8 installed\n"
+        f"\t* `onnx-jp6-cu126` - for running on Jetson with Jetpack 6\n"
+        f"If you see this error using Roboflow infrastructure, make sure the service you use does support the model. "
+        f"You can also contact Roboflow to get support.",
+        help_url="https://todo",
+    ) from import_error
+
+
+class YOLOACTForInstanceSegmentationOnnx(
+    InstanceSegmentationModel[
+        torch.Tensor,
+        PreProcessingMetadata,
+        Tuple[torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor],
+    ]
+):
+
+    @classmethod
+    def from_pretrained(
+        cls,
+        model_name_or_path: str,
+        onnx_execution_providers: Optional[List[Union[str, tuple]]] = None,
+        default_onnx_trt_options: bool = True,
+        device: torch.device = DEFAULT_DEVICE,
+        **kwargs,
+    ) -> "YOLOACTForInstanceSegmentationOnnx":
+        if onnx_execution_providers is None:
+            onnx_execution_providers = get_selected_onnx_execution_providers()
+        if not onnx_execution_providers:
+            raise EnvironmentConfigurationError(
+                message=f"Could not initialize model - selected backend is ONNX which requires execution provider to "
+                f"be specified - explicitly in `from_pretrained(...)` method or via env variable "
+                f"`ONNXRUNTIME_EXECUTION_PROVIDERS`. If you run model locally - adjust your setup, otherwise "
+                f"contact the platform support.",
+                help_url="https://todo",
+            )
+        onnx_execution_providers = set_execution_provider_defaults(
+            providers=onnx_execution_providers,
+            model_package_path=model_name_or_path,
+            device=device,
+            default_onnx_trt_options=default_onnx_trt_options,
+        )
+        model_package_content = get_model_package_contents(
+            model_package_dir=model_name_or_path,
+            elements=[
+                "class_names.txt",
+                "inference_config.json",
+                "weights.onnx",
+            ],
+        )
+        class_names = parse_class_names_file(
+            class_names_path=model_package_content["class_names.txt"]
+        )
+        inference_config = parse_inference_config(
+            config_path=model_package_content["inference_config.json"],
+            allowed_resize_modes={
+                ResizeMode.STRETCH_TO,
+                ResizeMode.LETTERBOX,
+                ResizeMode.CENTER_CROP,
+                ResizeMode.LETTERBOX_REFLECT_EDGES,
+            },
+        )
+        session = onnxruntime.InferenceSession(
+            path_or_bytes=model_package_content["weights.onnx"],
+            providers=onnx_execution_providers,
+        )
+        input_batch_size = session.get_inputs()[0].shape[0]
+        if input_batch_size != 1:
+            raise ModelRuntimeError(
+                message="Implementation of YOLOACTForInstanceSegmentationOnnx is adjusted to work correctly with "
+                "onnx models accepting inputs with `batch_size=1`. It can be extended if needed, but we've "
+                "not heard such request so far. If you find that a valueble feature - let us know via "
+                "https://github.com/roboflow/inference/issues"
+            )
+        input_name = session.get_inputs()[0].name
+        return cls(
+            session=session,
+            input_name=input_name,
+            class_names=class_names,
+            inference_config=inference_config,
+            device=device,
+        )
+
+    def __init__(
+        self,
+        session: onnxruntime.InferenceSession,
+        input_name: str,
+        inference_config: InferenceConfig,
+        class_names: List[str],
+        device: torch.device,
+    ):
+        self._session = session
+        self._input_name = input_name
+        self._inference_config = inference_config
+        self._class_names = class_names
+        self._device = device
+        self._session_thread_lock = Lock()
+
+    @property
+    def class_names(self) -> List[str]:
+        return self._class_names
+
+    def pre_process(
+        self,
+        images: Union[torch.Tensor, List[torch.Tensor], np.ndarray, List[np.ndarray]],
+        input_color_format: Optional[ColorFormat] = None,
+        **kwargs,
+    ) -> Tuple[torch.Tensor, List[PreProcessingMetadata]]:
+        return pre_process_network_input(
+            images=images,
+            image_pre_processing=self._inference_config.image_pre_processing,
+            network_input=self._inference_config.network_input,
+            target_device=self._device,
+            input_color_format=input_color_format,
+        )
+
+    def forward(
+        self, pre_processed_images: torch.Tensor, **kwargs
+    ) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor]:
+        with self._session_thread_lock:
+            (
+                all_loc_data,
+                all_conf_data,
+                all_mask_data,
+                all_prior_data,
+                all_proto_data,
+            ) = ([], [], [], [], [])
+            for image in pre_processed_images:
+                loc_data, conf_data, mask_data, prior_data, proto_data = (
+                    run_session_with_batch_size_limit(
+                        session=self._session,
+                        inputs={self._input_name: image.unsqueeze(0).contiguous()},
+                    )
+                )
+                all_loc_data.append(loc_data)
+                all_conf_data.append(conf_data)
+                all_mask_data.append(mask_data)
+                all_prior_data.append(prior_data)
+                all_proto_data.append(proto_data)
+            return (
+                torch.cat(all_loc_data, dim=0),
+                torch.cat(all_conf_data, dim=0),
+                torch.cat(all_mask_data, dim=0),
+                torch.stack(all_prior_data, dim=0),
+                torch.cat(all_proto_data, dim=0),
+            )
+
+    def post_process(
+        self,
+        model_results: Tuple[
+            torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor
+        ],
+        pre_processing_meta: List[PreProcessingMetadata],
+        conf_thresh: float = 0.25,
+        iou_thresh: float = 0.45,
+        max_detections: int = 100,
+        class_agnostic: bool = False,
+        **kwargs,
+    ) -> List[InstanceDetections]:
+        all_loc_data, all_conf_data, all_mask_data, all_prior_data, all_proto_data = (
+            model_results
+        )
+        batch_size = all_loc_data.shape[0]
+        num_priors = all_loc_data.shape[1]
+        boxes = torch.zeros((batch_size, num_priors, 4), device=self._device)
+        for batch_element_id, (
+            batch_element_loc_data,
+            batch_element_priors,
+            image_prep_meta,
+        ) in enumerate(zip(all_loc_data, all_prior_data, pre_processing_meta)):
+            image_boxes = decode_predicted_bboxes(
+                loc_data=batch_element_loc_data,
+                priors=batch_element_priors,
+            )
+            inference_height, inference_width = (
+                image_prep_meta.inference_size.height,
+                image_prep_meta.inference_size.width,
+            )
+            scale = torch.tensor(
+                [inference_width, inference_height, inference_width, inference_height],
+                device=self._device,
+            )
+            image_boxes = image_boxes.mul_(scale)
+            boxes[batch_element_id, :, :] = image_boxes
+        all_conf_data = all_conf_data[:, :, 1:]  # remove background class
+        instances = torch.cat([boxes, all_conf_data, all_mask_data], dim=2)
+        nms_results = run_nms_for_instance_segmentation(
+            output=instances,
+            conf_thresh=conf_thresh,
+            iou_thresh=iou_thresh,
+            max_detections=max_detections,
+            class_agnostic=class_agnostic,
+        )
+        final_results = []
+        for image_bboxes, image_protos, image_meta in zip(
+            nms_results, all_proto_data, pre_processing_meta
+        ):
+            pre_processed_masks = image_protos @ image_bboxes[:, 6:].T
+            pre_processed_masks = 1 / (1 + torch.exp(-pre_processed_masks))
+            pre_processed_masks = torch.permute(pre_processed_masks, (2, 0, 1))
+            cropped_masks = crop_masks_to_boxes(
+                image_bboxes[:, :4], pre_processed_masks
+            )
+            padding = (
+                image_meta.pad_left,
+                image_meta.pad_top,
+                image_meta.pad_right,
+                image_meta.pad_bottom,
+            )
+            aligned_boxes, aligned_masks = align_instance_segmentation_results(
+                image_bboxes=image_bboxes,
+                masks=cropped_masks,
+                padding=padding,
+                scale_height=image_meta.scale_height,
+                scale_width=image_meta.scale_width,
+                original_size=image_meta.original_size,
+                size_after_pre_processing=image_meta.size_after_pre_processing,
+                inference_size=image_meta.inference_size,
+                static_crop_offset=image_meta.static_crop_offset,
+                binarization_threshold=0.5,
+            )
+            final_results.append(
+                InstanceDetections(
+                    xyxy=aligned_boxes[:, :4].round().int(),
+                    class_id=aligned_boxes[:, 5].int(),
+                    confidence=aligned_boxes[:, 4],
+                    mask=aligned_masks,
+                )
+            )
+        return final_results
+
+
+def decode_predicted_bboxes(
+    loc_data: torch.Tensor, priors: torch.Tensor
+) -> torch.Tensor:
+    variances = torch.tensor([0.1, 0.2], device=loc_data.device)
+    boxes = torch.cat(
+        [
+            priors[:, :2] + loc_data[:, :2] * variances[0] * priors[:, 2:],
+            priors[:, 2:] * torch.exp(loc_data[:, 2:] * variances[1]),
+        ],
+        dim=1,
+    )
+    boxes[:, :2] -= boxes[:, 2:] / 2
+    boxes[:, 2:] += boxes[:, :2]
+    return boxes
+
+
+def run_nms_for_instance_segmentation(
+    output: torch.Tensor,
+    conf_thresh: float = 0.25,
+    iou_thresh: float = 0.45,
+    max_detections: int = 100,
+    class_agnostic: bool = False,
+) -> List[torch.Tensor]:
+    bs = output.shape[0]
+    boxes = output[:, :, :4]  # (N, 19248, 4)
+    scores = output[:, :, 4:-32]  # (N, 19248, num_classes)
+    masks = output[:, :, -32:]
+    results = []
+    for b in range(bs):
+        bboxes = boxes[b]  # (19248, 4)
+        class_scores = scores[b]  # (19248, 80)
+        box_masks = masks[b]
+        class_conf, class_ids = class_scores.max(1)  # (8400,), (8400,)
+        mask = class_conf > conf_thresh
+        if mask.sum() == 0:
+            results.append(torch.zeros((0, 38), device=output.device))
+            continue
+        bboxes = bboxes[mask]
+        class_conf = class_conf[mask]
+        class_ids = class_ids[mask]
+        box_masks = box_masks[mask]
+        # Class-agnostic NMS -> use dummy class ids
+        nms_class_ids = torch.zeros_like(class_ids) if class_agnostic else class_ids
+        keep = torchvision.ops.batched_nms(
+            bboxes, class_conf, nms_class_ids, iou_thresh
+        )
+        keep = keep[:max_detections]
+        detections = torch.cat(
+            [
+                bboxes[keep],
+                class_conf[keep].unsqueeze(1),
+                class_ids[keep].unsqueeze(1).float(),
+                box_masks[keep],
+            ],
+            dim=1,
+        )  # [x1, y1, x2, y2, conf, cls]
+        results.append(detections)
+    return results