update for transformers compatible

.gitignore

@@ -1 +0,0 @@
-*.py

README.md

@@ -5,6 +5,7 @@ license: apache-2.0
 tags:
     - object-detection
     - AgTech
+    - transformers
 library_name: pytorch
 inference: false
 datasets:

config.json

@@ -0,0 +1,13 @@
+{
+    "model_type": "dfine",
+    "architectures": ["DFineModel"],
+    "auto_map": {
+      "AutoConfig": "d-fine-large--configuration_dfine.DFineConfig",
+      "AutoModel": "d-fine-large--modeling_dfine.DFineModel"
+    },
+    "torch_dtype": "float32",
+    "transformers_version": "4.51.3",
+    "input_size": [640, 640],
+    "input_components": ["images", "orig_target_sizes", "ratio", "pad_w", "pad_h"],
+    "output_components": ["labels", "boxes", "scores"]
+}

config.yaml

@@ -1,41 +0,0 @@
-task: detection
-
-evaluator:
-  type: CocoEvaluator
-  iou_types: ['bbox', ]
-
-num_classes: 3 # your dataset classes
-remap_mscoco_category: False
-
-train_dataloader:
-  type: DataLoader
-  dataset:
-    type: CocoDetection
-    img_folder: /dataset/images/train
-    ann_file: /dataset/annotations/instances_train.json
-    return_masks: False
-    transforms:
-      type: Compose
-      ops: ~
-  shuffle: True
-  num_workers: 4
-  drop_last: True
-  collate_fn:
-    type: BatchImageCollateFunction
-
-
-val_dataloader:
-  type: DataLoader
-  dataset:
-    type: CocoDetection
-    img_folder: /dataset/images/val
-    ann_file: /dataset/annotations/instances_val.json
-    return_masks: False
-    transforms:
-      type: Compose
-      ops: ~
-  shuffle: False
-  num_workers: 4
-  drop_last: False
-  collate_fn:
-    type: BatchImageCollateFunction

configuration_dfine.py

@@ -0,0 +1,16 @@
+from transformers import PretrainedConfig
+
+class DFineConfig(PretrainedConfig):
+    model_type = "dfine"
+
+    def __init__(
+        self,
+        input_size=[640, 640],
+        input_components=["images", "orig_target_sizes"],
+        output_components=["labels", "boxes", "scores"],
+        **kwargs
+    ):
+        super().__init__(**kwargs)
+        self.input_size = input_size
+        self.input_components = input_components
+        self.output_components = output_components

model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa29629571161f24e347e7190ee10432dbd4a1787f01893f84beea9fc00584da
+size 123992203

modeling_dfine.py

@@ -0,0 +1,51 @@
+import os
+import torch
+import onnxruntime as ort
+from huggingface_hub import hf_hub_download
+from transformers import PreTrainedModel
+from .configuration_dfine import DFineConfig
+
+class DFineModel(PreTrainedModel):
+    config_class = DFineConfig
+
+    def __init__(self, config):
+        super().__init__(config)
+        model_path = hf_hub_download(
+            repo_id="Laudando-Associates-LLC/d-fine-large",
+            filename="model.onnx"
+        )
+        self.session = ort.InferenceSession(model_path, providers=["CPUExecutionProvider"])
+
+    def forward(self, images, orig_target_sizes, ratio, pad_w, pad_h, conf_threshold=0.5):
+        output = self.session.run(
+            output_names=None,
+            input_feed={"images": images.numpy(), "orig_target_sizes": orig_target_sizes.numpy()},
+        )
+        labels, boxes, scores = output
+
+        # Convert to torch
+        labels = torch.tensor(labels)
+        boxes = torch.tensor(boxes)
+        scores = torch.tensor(scores)
+
+        # Filter by confidence per image
+        results = []
+        for i in range(scores.shape[0]):
+            keep = scores[i] > conf_threshold
+            labels_kept = labels[i][keep]
+            boxes_kept = boxes[i][keep]
+            scores_kept = scores[i][keep]
+
+            # Auto-scale boxes back to original image resolution
+            boxes_scaled = boxes_kept.clone()
+            boxes_scaled[:, 0::2] -= pad_w[i]
+            boxes_scaled[:, 1::2] -= pad_h[i]
+            boxes_scaled /= ratio[i]
+
+            results.append({
+                "labels": labels_kept,
+                "boxes": boxes_scaled,
+                "scores": scores_kept
+            })
+
+        return results

requirements.txt

@@ -0,0 +1 @@
+onnxruntime