Upload 6 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+PP-DocLayoutV3.onnx.data filter=lfs diff=lfs merge=lfs -text

PP-DocLayoutV3.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0721928ff08741bb208ebed539c77170db5234a68cb7e546e6cc9bc172a695b
+size 5088167

PP-DocLayoutV3.onnx.data

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34df3e4b79d7bbbf82abce1b4f3cde3d540fa57ad42ec8905c352b97c408d437
+size 136774480

README.md

@@ -1,3 +1,110 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+library_name: onnx
+tags:
+  - onnx
+  - object-detection
+  - layout-analysis
+  - document-understanding
+  - paddleocr
+base_model: PaddlePaddle/PP-DocLayoutV3_safetensors
+---
+
+# PP-DocLayoutV3 — ONNX export
+
+ONNX export of [PaddlePaddle/PP-DocLayoutV3_safetensors](https://huggingface.co/PaddlePaddle/PP-DocLayoutV3_safetensors), the layout-detection model used in the PaddleOCR-VL-1.5 pipeline.
+
+This export preserves all four model heads: classification logits, bounding boxes, instance-segmentation masks, and reading-order logits. The original PaddlePaddle release outputs polygon points and reading order via a postprocessor that consumes these four tensors.
+
+## Files
+
+| file | size | purpose |
+|---|---|---|
+| `PP-DocLayoutV3.onnx` | ~5 MB | model graph (references external weights) |
+| `PP-DocLayoutV3.onnx.data` | ~137 MB | weight tensors (must sit alongside `.onnx`) |
+| `config.json` | — | original model config (HuggingFace-style) |
+| `preprocessor_config.json` | — | image preprocessing parameters (800×800 resize, normalize) |
+| `inference.yml` | — | original PaddlePaddle inference config (reference only) |
+
+## Inputs / outputs
+
+**Input** (single tensor):
+
+| name | shape | dtype | notes |
+|---|---|---|---|
+| `pixel_values` | `(B, 3, 800, 800)` | `float32` | Resize image to 800×800, rescale by `1/255`, mean=`[0,0,0]`, std=`[1,1,1]` (matches `preprocessor_config.json`). |
+
+**Outputs** (four tensors):
+
+| name | shape | notes |
+|---|---|---|
+| `logits` | `(B, 300, 25)` | per-query class logits over 25 layout classes |
+| `pred_boxes` | `(B, 300, 4)` | normalized `(cx, cy, w, h)` — convert via standard DETR decoding |
+| `out_masks` | `(B, 300, 200, 200)` | per-query instance-segmentation masks; cv2 contour extraction yields polygon points |
+| `order_logits` | `(B, 300, 300)` | per-query permutation logits for reading order; argmax / Sinkhorn for ordering |
+
+## Postprocessing
+
+The official postprocessor lives in `transformers.models.pp_doclayout_v3.image_processing_pp_doclayout_v3.PPDocLayoutV3ImageProcessor.post_process_object_detection`. It takes the four output tensors plus a `target_sizes` tensor and returns:
+
+```
+{
+  "scores":         (N,)      float32
+  "labels":         (N,)      int64
+  "boxes":          (N, 4)    float32 — axis-aligned (x1, y1, x2, y2) in target coords
+  "polygon_points": list[N]   each (P, 2) int polygon vertices in target coords
+  "order_seq":      (N,)      int64   — reading-order index
+}
+```
+
+You can use that postprocessor directly (`transformers >= 5.4`, requires `torch` and `cv2`) or port it to numpy + cv2 for a torch-free runtime.
+
+## Loading
+
+```python
+import onnxruntime as ort
+import numpy as np
+
+sess = ort.InferenceSession("PP-DocLayoutV3.onnx", providers=["CPUExecutionProvider"])
+# preprocess to 800x800 RGB float32, normalize per preprocessor_config.json
+pixel_values = ...  # shape (1, 3, 800, 800), float32
+logits, pred_boxes, out_masks, order_logits = sess.run(
+    ["logits", "pred_boxes", "out_masks", "order_logits"],
+    {"pixel_values": pixel_values},
+)
+```
+
+The `.onnx.data` sidecar is loaded automatically by onnxruntime via the relative `location` reference embedded in the graph. Both files must sit in the same directory.
+
+## How this was exported
+
+1. `pip install transformers==5.6.2 torch==2.11 onnx==1.21 onnxscript`
+2. `model = AutoModelForObjectDetection.from_pretrained("PaddlePaddle/PP-DocLayoutV3_safetensors").eval()`
+3. Wrap the model so `forward(pixel_values)` returns `(logits, pred_boxes, out_masks, order_logits)`.
+4. `torch.onnx.export(wrapped, (pixel_values,), "PP-DocLayoutV3.onnx", opset_version=18, dynamo=True, dynamic_axes={"pixel_values": {0: "batch"}})`
+5. Re-save with `onnx.save(..., save_as_external_data=True, location="PP-DocLayoutV3.onnx.data")` to standardize the sidecar filename.
+
+Numerical parity vs torch (random `(1, 3, 800, 800)` input):
+
+| output | max absolute diff |
+|---|---|
+| `logits` | 1.32e-4 |
+| `pred_boxes` | 1.57e-5 |
+| `out_masks` | 1.62e-3 |
+| `order_logits` | 3.96e-2 |
+
+The `order_logits` deviation reflects accumulated floating-point drift in the decoder's attention; argmax-based reading order is unaffected on the test images we checked.
+
+## Inference speed
+
+CPU (Apple M-series, single page, 800×800 input): **~480 ms/page** with `CPUExecutionProvider`.
+
+## Source
+
+- Original weights: [PaddlePaddle/PP-DocLayoutV3_safetensors](https://huggingface.co/PaddlePaddle/PP-DocLayoutV3_safetensors)
+- Original PaddlePaddle release: [PaddlePaddle/PP-DocLayoutV3](https://huggingface.co/PaddlePaddle/PP-DocLayoutV3)
+- Paper: [PaddleOCR-VL-1.5 (arXiv:2601.21957)](https://arxiv.org/abs/2601.21957)
+
+## License
+
+Apache-2.0 (inherited from PaddlePaddle/PP-DocLayoutV3_safetensors).

config.json

@@ -0,0 +1,107 @@
+{
+  "activation_dropout": 0.0,
+  "activation_function": "silu",
+  "anchor_image_size": null,
+  "architectures": [
+    "PPDocLayoutV3ForObjectDetection"
+  ],
+  "attention_dropout": 0.0,
+  "backbone": null,
+  "backbone_config": {
+    "model_type": "hgnet_v2",
+    "arch": "L",
+    "return_idx": [0, 1, 2, 3],
+    "freeze_stem_only": true,
+    "freeze_at": 0,
+    "freeze_norm": true,
+    "lr_mult_list": [0, 0.05, 0.05, 0.05, 0.05],
+    "out_features": ["stage1", "stage2", "stage3", "stage4"]
+  },
+  "backbone_kwargs": null,
+  "batch_norm_eps": 1e-05,
+  "box_noise_scale": 1.0,
+  "d_model": 256,
+  "decoder_activation_function": "relu",
+  "decoder_attention_heads": 8,
+  "decoder_ffn_dim": 1024,
+  "decoder_in_channels": [
+    256,
+    256,
+    256
+  ],
+  "decoder_layers": 6,
+  "decoder_n_points": 4,
+  "disable_custom_kernels": true,
+  "dropout": 0.0,
+  "encode_proj_layers": [
+    2
+  ],
+  "encoder_activation_function": "gelu",
+  "encoder_attention_heads": 8,
+  "encoder_ffn_dim": 1024,
+  "encoder_hidden_dim": 256,
+  "encoder_in_channels": [
+    512,
+    1024,
+    2048
+  ],
+  "encoder_layers": 1,
+  "eos_coefficient": 0.0001,
+  "eval_size": null,
+  "feature_strides": [
+    8,
+    16,
+    32
+  ],
+  "hidden_expansion": 1.0,
+  "id2label": {
+    "0": "abstract",
+    "1": "algorithm",
+    "2": "aside_text",
+    "3": "chart",
+    "4": "content",
+    "5": "formula",
+    "6": "doc_title",
+    "7": "figure_title",
+    "8": "footer",
+    "9": "footer",
+    "10": "footnote",
+    "11": "formula_number",
+    "12": "header",
+    "13": "header",
+    "14": "image",
+    "15": "formula",
+    "16": "number",
+    "17": "paragraph_title",
+    "18": "reference",
+    "19": "reference_content",
+    "20": "seal",
+    "21": "table",
+    "22": "text",
+    "23": "text",
+    "24": "vision_footnote"
+  },
+  "initializer_range": 0.01,
+  "is_encoder_decoder": true,
+  "label2id": {},
+  "label_noise_ratio": 0.5,
+  "layer_norm_eps": 1e-05,
+  "learn_initial_query": false,
+  "matcher_alpha": 0.25,
+  "matcher_bbox_cost": 5.0,
+  "matcher_class_cost": 2.0,
+  "matcher_gamma": 2.0,
+  "matcher_giou_cost": 2.0,
+  "model_type": "pp_doclayout_v3",
+  "normalize_before": false,
+  "num_denoising": 100,
+  "num_feature_levels": 3,
+  "num_queries": 300,
+  "positional_encoding_temperature": 10000,
+  "torch_dtype": "float32",
+  "use_pretrained_backbone": false,
+  "use_timm_backbone": false,
+  "global_pointer_head_size": 64,
+  "mask_feature_channels": [64, 64],
+  "x4_feat_dim": 128
+}

inference.yml

@@ -0,0 +1,100 @@
+mode: paddle
+draw_threshold: 0.5
+metric: COCO
+use_dynamic_shape: false
+Global:
+  model_name: PP-DocLayoutV3
+arch: DETR
+min_subgraph_size: 3
+Preprocess:
+- interp: 2
+  keep_ratio: false
+  target_size:
+  - 800
+  - 800
+  type: Resize
+- mean:
+  - 0.0
+  - 0.0
+  - 0.0
+  norm_type: none
+  std:
+  - 1.0
+  - 1.0
+  - 1.0
+  type: NormalizeImage
+- type: Permute
+label_list:
+- abstract
+- algorithm
+- aside_text
+- chart
+- content
+- display_formula
+- doc_title
+- figure_title
+- footer
+- footer_image
+- footnote
+- formula_number
+- header
+- header_image
+- image
+- inline_formula
+- number
+- paragraph_title
+- reference
+- reference_content
+- seal
+- table
+- text
+- vertical_text
+- vision_footnote
+Hpi:
+  backend_configs:
+    paddle_infer:
+      trt_dynamic_shapes: &id001
+        image:
+        - - 1
+          - 3
+          - 800
+          - 800
+        - - 1
+          - 3
+          - 800
+          - 800
+        - - 8
+          - 3
+          - 800
+          - 800
+        scale_factor:
+        - - 1
+          - 2
+        - - 1
+          - 2
+        - - 8
+          - 2
+      trt_dynamic_shape_input_data:
+        scale_factor:
+        - - 2
+          - 2
+        - - 1
+          - 1
+        - - 0.67
+          - 0.67
+          - 0.67
+          - 0.67
+          - 0.67
+          - 0.67
+          - 0.67
+          - 0.67
+          - 0.67
+          - 0.67
+          - 0.67
+          - 0.67
+          - 0.67
+          - 0.67
+          - 0.67
+          - 0.67
+    tensorrt:
+      dynamic_shapes: *id001

preprocessor_config.json

@@ -0,0 +1,36 @@
+{
+  "_valid_processor_keys": [
+    "images",
+    "do_resize",
+    "size",
+    "resample",
+    "do_rescale",
+    "rescale_factor",
+    "do_normalize",
+    "image_mean",
+    "image_std",
+    "return_tensors",
+    "data_format",
+    "input_data_format"
+  ],
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0,
+    0,
+    0
+  ],
+  "image_processor_type": "PPDocLayoutV3ImageProcessor",
+  "image_std": [
+    1,
+    1,
+    1
+  ],
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 800,
+    "width": 800
+  }
+}