Upload folder using huggingface_hub

README.md

@@ -0,0 +1,92 @@
+---
+pipeline_tag: object-detection
+license: apache-2.0
+base_model: microsoft/table-transformer-structure-recognition
+library_name: zeromodels
+tags:
+- keras
+- zeromodels
+- table-transformer
+- detr
+- object-detection
+- arxiv:2110.00061
+- pytorch
+- jax
+- tf
+---
+
+## ***See [our collection](https://huggingface.co/collections/zeromodels/table-transformer-6a928f000a88c73a41f9411e) for all versions of Table Transformer.***
+
+# Run Table Transformer with Keras 3: JAX, PyTorch, or TensorFlow
+
+[![GitHub](https://img.shields.io/badge/GitHub-ZeroModels-black?logo=github)](https://github.com/IMvision12/ZeroModels) [![Docs](https://img.shields.io/badge/Docs-Table%20Transformer-blue)](https://imvision12.github.io/ZeroModels/table_transformer/) [![Collection](https://img.shields.io/badge/HF-Table%20Transformer%20collection-yellow)](https://huggingface.co/collections/zeromodels/table-transformer-6a928f000a88c73a41f9411e)
+
+# zeromodels/table-transformer-structure-recognition
+
+Paper: [PubTables-1M: Towards comprehensive table extraction from unstructured documents (arXiv:2110.00061)](https://arxiv.org/abs/2110.00061) · [HF Papers](https://huggingface.co/papers/2110.00061)
+
+Table Transformer (TATR) applies the DETR detection recipe to tables. A ResNet-18 backbone produces a feature map, a transformer encoder-decoder attends over it with a fixed set of learned object queries, and each query emits one class and one box, with no anchors and no NMS. This structure-recognition checkpoint decomposes a cropped table into its rows, columns, and header / spanning cells.
+
+For more details on the model, please go to Microsoft's original [model card](https://huggingface.co/microsoft/table-transformer-structure-recognition).
+
+Pure-**Keras 3** conversion of [`microsoft/table-transformer-structure-recognition`](https://huggingface.co/microsoft/table-transformer-structure-recognition) for [zeromodels](https://github.com/IMvision12/ZeroModels). One implementation runs unmodified on **TensorFlow / Torch / JAX**.
+
+This is a **table-structure recognition** checkpoint (`TableTransformerDetect`): each query predicts a table element (classes: table, column, row, column header, projected row header, spanning cell).
+
+## ✨ Quick start
+
+```python
+import os
+os.environ["KERAS_BACKEND"] = "torch"  # or "jax" / "tensorflow"
+
+from PIL import Image
+from zeromodels.models.table_transformer import (
+    TableTransformerDetect,
+    TableTransformerImageProcessor,
+)
+from zeromodels.models.table_transformer.table_transformer_image_processor import (
+    TABLE_STRUCTURE_LABELS,
+)
+
+model = TableTransformerDetect.from_weights("zeromodels/table-transformer-structure-recognition")
+processor = TableTransformerImageProcessor.from_weights("zeromodels/table-transformer-structure-recognition")
+
+image = Image.open("your_image.jpg").convert("RGB")
+inputs = processor(image)
+output = model(inputs["pixel_values"], training=False)
+results = processor.post_process_object_detection(
+    output,
+    threshold=0.6,
+    target_sizes=[(image.height, image.width)],
+    label_names=TABLE_STRUCTURE_LABELS,
+)[0]
+for score, name, box in zip(
+    results["scores"], results["label_names"], results["boxes"]
+):
+    print(f"{name}: {float(score):.3f} {box}")
+```
+
+Load any Table Transformer variant the same way with `from_weights("zeromodels/<variant>")`:
+
+| Variant | Hub | Task |
+|---|---|---|
+| `table-transformer-detection` | [`zeromodels/table-transformer-detection`](https://huggingface.co/zeromodels/table-transformer-detection) | table detection |
+| `table-transformer-structure-recognition` | [`zeromodels/table-transformer-structure-recognition`](https://huggingface.co/zeromodels/table-transformer-structure-recognition) | table-structure recognition |
+| `table-transformer-structure-recognition-v1.1-all` | [`zeromodels/table-transformer-structure-recognition-v1.1-all`](https://huggingface.co/zeromodels/table-transformer-structure-recognition-v1.1-all) | table-structure recognition |
+| `table-transformer-structure-recognition-v1.1-fin` | [`zeromodels/table-transformer-structure-recognition-v1.1-fin`](https://huggingface.co/zeromodels/table-transformer-structure-recognition-v1.1-fin) | table-structure recognition |
+| `table-transformer-structure-recognition-v1.1-pub` | [`zeromodels/table-transformer-structure-recognition-v1.1-pub`](https://huggingface.co/zeromodels/table-transformer-structure-recognition-v1.1-pub) | table-structure recognition |
+
+## Tips
+
+- Set `KERAS_BACKEND` **before** importing Keras / zeromodels.
+- Detection finds tables in a page image; structure recognition decomposes a cropped table into rows, columns, and header / spanning cells.
+- Both tasks use `TableTransformerDetect` + `post_process_object_detection`; only the checkpoint and the class set differ.
+- Pass `label_names=TABLE_DETECTION_LABELS` for the detection checkpoint and `label_names=TABLE_STRUCTURE_LABELS` for structure recognition (both live in `zeromodels.models.table_transformer.table_transformer_image_processor`).
+- See [Table Transformer docs](https://imvision12.github.io/ZeroModels/table_transformer/) and [Loading Weights](https://imvision12.github.io/ZeroModels/loading_weights/).
+- Community / upstream safetensors still work via the `hf:` prefix, e.g. `TableTransformerDetect.from_weights("hf:microsoft/table-transformer-structure-recognition")`.
+
+## Special Thanks
+
+A huge thank you to the Microsoft Table Transformer authors for creating and releasing these models.
+
+License: Apache 2.0.

model.weights.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:005a19304c3f5bc6a1a207adf45eee8d7ee6c5f94beb39a74b191a49a8510be1
+size 116199352

zm_config.json

@@ -0,0 +1,21 @@
+{
+  "library_name": "zeromodels",
+  "zeromodels_version": "1.2.7",
+  "model_module": "zeromodels.models.table_transformer",
+  "model_class": "TableTransformerDetect",
+  "variant": "table-transformer-structure-recognition",
+  "weights": "model.weights.h5",
+  "schema_version": 2,
+  "model_type": "table-transformer",
+  "vision_config": {
+    "hidden_dim": 256,
+    "num_heads": 8,
+    "num_encoder_layers": 6,
+    "num_decoder_layers": 6,
+    "dim_feedforward": 2048,
+    "dropout_rate": 0.1,
+    "num_queries": 125,
+    "num_classes": 7,
+    "image_size": 800
+  }
+}

zm_preprocessor.json

@@ -0,0 +1,27 @@
+{
+  "library_name": "zeromodels",
+  "zeromodels_version": "1.2.7",
+  "preprocessor_module": "zeromodels.models.table_transformer",
+  "preprocessor_class": "TableTransformerImageProcessor",
+  "variant": "table-transformer-structure-recognition",
+  "size": {
+    "height": 800,
+    "width": 800
+  },
+  "resample": "bilinear",
+  "do_rescale": true,
+  "rescale_factor": 0.00392156862745098,
+  "do_normalize": true,
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "return_tensor": true,
+  "data_format": null
+}