Consolidate the FleXray ensemble into one repository (members/ + ensemble.json)

README.md

@@ -12,7 +12,7 @@ tags:
   - flexray
 ---
 
-# FleXray: `flexray-base`
+# FleXray: `VictorButoi/flexray`
 
 - Website and in-browser demo: [victorbutoi.github.io/FleXray](https://victorbutoi.github.io/FleXray/)
 - Code: [github.com/VictorButoi/FleXray](https://github.com/VictorButoi/FleXray)
@@ -20,16 +20,15 @@ tags:
 - Tutorial: [Colab notebook](https://colab.research.google.com/drive/1MNIeN9LN-tY8wAifWSolxmFZi3Q0BlVX)
 - Paper: *FleXray: Flexible Full-Body X-ray Segmentation* (coming soon)
 
-`flexray-base` is the **flagship FleXray model**: a single 2D UNet that segments
-anatomy from standard radiographs across body regions, projections, and
-acquisition settings. It predicts 60 anatomical structures (plus background) as
-independent sigmoid channels at 256 x 256 resolution and is the default bundle
-loaded by `flexify` and `FleXraySegmenter.from_pretrained`.
+FleXray is a single 2D UNet that segments anatomy from standard radiographs
+across body regions, projections, and acquisition settings. It predicts 60
+anatomical structures (plus background) as independent sigmoid channels at
+256 x 256 resolution.
 
-It is also one member of a five-model ensemble; see
-[The FleXray ensemble](#the-flexray-ensemble) for the other four and
-[Test-time augmentation](#test-time-augmentation) for the inference recipe we
-use for reported results.
+This repository holds the **flagship model** and the four sibling models of
+the FleXray ensemble, one bundle per `members/` subfolder (see
+[Repository layout](#repository-layout)). `flexify` and
+`FleXraySegmenter.from_pretrained` load the flagship by default.
 
 ## Quick start
 
@@ -41,7 +40,7 @@ flexify --input ./image.png --output-dir ./predictions
 ```python
 from fxr.inference import FleXraySegmenter
 
-segmenter = FleXraySegmenter.from_pretrained("VictorButoi/flexray-base")
+segmenter = FleXraySegmenter.from_pretrained("VictorButoi/flexray")
 prediction = segmenter.predict("./image.png", threshold=0.5)
 prediction.masks          # uint8, CxHxW thresholded masks
 prediction.probabilities  # float32, CxHxW sigmoid probabilities
@@ -60,41 +59,36 @@ The flagship was trained with a 0.375 FluXray proportion in the training mix.
 Four sibling models share its architecture, label schema, preprocessing, and
 training recipe and differ only in that proportion:
 
-| Model ID | FluXray proportion | Role |
+| Subfolder | FluXray proportion | Role |
 | --- | --- | --- |
-| `VictorButoi/flexray-base-flux000` | 0.0 | ensemble member |
-| `VictorButoi/flexray-base-flux025` | 0.25 | ensemble member |
-| **`VictorButoi/flexray-base`** | **0.375** | **flagship (this repository)** |
-| `VictorButoi/flexray-base-flux050` | 0.5 | ensemble member |
-| `VictorButoi/flexray-base-flux075` | 0.75 | ensemble member |
+| `members/flux000` | 0.0 | ensemble member |
+| `members/flux025` | 0.25 | ensemble member |
+| **`members/flux0375`** | **0.375** | **flagship (loaded by default)** |
+| `members/flux050` | 0.5 | ensemble member |
+| `members/flux075` | 0.75 | ensemble member |
 
-Because the members share one output space, any subset can be averaged in
-probability space:
+`ensemble.json` at the repository root lists the flagship and the members.
+Because the members share one output space, they are averaged in probability
+space:
 
 ```bash
-flexify --model-id VictorButoi/flexray-base \
-        --model-id VictorButoi/flexray-base-flux000 \
-        --model-id VictorButoi/flexray-base-flux025 \
-        --model-id VictorButoi/flexray-base-flux050 \
-        --model-id VictorButoi/flexray-base-flux075 \
-        --tta-samples 16 --input ./image.png --output-dir ./predictions
+flexify --ensemble --tta-samples 16 --input ./image.png --output-dir ./predictions
+flexify --subfolder members/flux000 --input ./image.png --output-dir ./predictions
 ```
 
 ```python
-segmenter = FleXraySegmenter.from_pretrained([
-    "VictorButoi/flexray-base",
-    "VictorButoi/flexray-base-flux000",
-    "VictorButoi/flexray-base-flux025",
-    "VictorButoi/flexray-base-flux050",
-    "VictorButoi/flexray-base-flux075",
-])
+segmenter = FleXraySegmenter.from_pretrained("VictorButoi/flexray", ensemble=True)
 prediction = segmenter.predict("./image.png", tta_samples=16)
+
+member = FleXraySegmenter.from_pretrained(
+    "VictorButoi/flexray", subfolder="members/flux000"
+)
 ```
 
 The website demo exposes the same choices as quality modes: **Low** runs the
 flagship once, **Normal** runs the flagship with 16-pass TTA, **High** runs the
 five-model ensemble once, and **X-High** runs the ensemble with 16-pass TTA.
-Release status of each member is tracked in
+The members are also listed in
 [MODEL_ZOO.md](https://github.com/VictorButoi/FleXray/blob/main/MODEL_ZOO.md).
 
 ## Test-time augmentation
@@ -145,7 +139,7 @@ FleXray outputs 60 foreground masks (61 channels including `background`). The
 broader dataset/evaluation protocol also recognizes aggregate `lumbar_spine` and
 `thoracolumbar_spine` annotations; these are evaluated by combining the relevant
 per-vertebra outputs and are not checkpoint channels. Channel order is stored in
-`label_schema.json`.
+each bundle's `label_schema.json`.
 
 - **Skull / shoulder girdle:** skull, scapulae, clavicles
 - **Upper limb:** humeri, radii, ulnae, carpals, metacarpals, phalanges
@@ -163,11 +157,11 @@ laterality is not predicted.
 `fxr.models.UNet`, 2D, 1 input channel, 61 output channels; filters
 `[64, 128, 256, 512, 512, 720, 1024]`, 3 convolutions per block, residual
 blocks with instance norm, `align_corners=True` upsampling. The full
-architecture is in `config.yml`.
+architecture is in each bundle's `config.yml`.
 
 ## Training data
 
-The model was trained on three source types unified under the FleXray label
+The models were trained on three source types unified under the FleXray label
 protocol. For training mixture proportions, please refer to the paper:
 
 - **Real X-ray masks:** HandBones, FootBones, MURA forearm, and MURA humerus,
@@ -188,8 +182,8 @@ Every dataset's license, redistribution status, and download pointer is
 documented in the
 [`VictorButoi/flexray-data`](https://huggingface.co/datasets/VictorButoi/flexray-data)
 card. That repository ships the real X-ray sources whose licenses permit
-redistribution, already packed in the FleXray protocol, plus the MURA masks;
-the FluXray database is released alongside it.
+redistribution, already packed in the FleXray protocol, the MURA masks, and
+the FluXray database.
 
 ## Evaluation
 
@@ -214,16 +208,20 @@ clinical diagnosis, treatment planning, or patient-care decisions.
 - Performance on acquisition settings, views, or populations far from the
   training sources has not been validated.
 
-## Files
-
-- `model.safetensors`: exported model weights.
-- `config.yml`: architecture and protocol config consumed by `from_pretrained`.
-- `label_schema.json`: ordered output labels.
-- `preprocessing.json`: public preprocessing contract.
-- `checksums.json`: SHA256 checksums of the bundle files.
-- `onnx/flexray-base-256-fp16.onnx`: fp16 ONNX export (opset 18, sigmoid
-  baked in) used by the in-browser demo; parity-checked against the PyTorch
-  weights by `tools/export_web_demo.py`.
+## Repository layout
+
+- `README.md`: this card.
+- `ensemble.json`: the `flagship` subfolder and the `members` list with their
+  FluXray proportions.
+- `members/<name>/model.safetensors`: exported model weights.
+- `members/<name>/config.yml`: architecture and protocol config consumed by
+  `from_pretrained`.
+- `members/<name>/label_schema.json`: ordered output labels.
+- `members/<name>/preprocessing.json`: public preprocessing contract.
+- `members/<name>/checksums.json`: SHA256 checksums of the bundle files.
+- `members/<name>/onnx/flexray-<name>-256-fp16.onnx`: fp16 ONNX export
+  (opset 18, sigmoid baked in) used by the in-browser demo; parity-checked
+  against the PyTorch weights by `tools/export_web_demo.py`.
 
 ## Licenses
 

ensemble.json

@@ -0,0 +1,26 @@
+{
+  "aggregation": "mean_probability",
+  "flagship": "members/flux0375",
+  "members": [
+    {
+      "fluxray_proportion": 0.375,
+      "subfolder": "members/flux0375"
+    },
+    {
+      "fluxray_proportion": 0.0,
+      "subfolder": "members/flux000"
+    },
+    {
+      "fluxray_proportion": 0.25,
+      "subfolder": "members/flux025"
+    },
+    {
+      "fluxray_proportion": 0.5,
+      "subfolder": "members/flux050"
+    },
+    {
+      "fluxray_proportion": 0.75,
+      "subfolder": "members/flux075"
+    }
+  ]
+}

members/flux000/checksums.json

@@ -0,0 +1,9 @@
+{
+  "algorithm": "sha256",
+  "files": {
+    "config.yml": "7c7d52b2404d5e3cb5347036feac994ab8b8a18dd6bef5cfa47b749077af727a",
+    "label_schema.json": "6f60e2369859b7e17e275c04394c9ec33ac194a8118111577886b1c7fd3cf9f5",
+    "model.safetensors": "921ffc5d146882b5912a4930c393ef540bc9f511bd0a0a4191fe10c3d951eaee",
+    "preprocessing.json": "5994ede07dc359e7fd62215ed1e97b811360fccacb6327bd91b6def439926155"
+  }
+}

label_schema.json → members/flux000/label_schema.json

@@ -63,7 +63,7 @@
     "heart"
   ],
   "label_order_verified": true,
-  "model_id": "VictorButoi/flexray-base",
+  "model_id": "VictorButoi/flexray",
   "num_labels": 61,
   "schema_version": 1,
   "weights_license": "CC-BY-NC-4.0"

members/flux000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:921ffc5d146882b5912a4930c393ef540bc9f511bd0a0a4191fe10c3d951eaee
+size 405630476

members/flux000/onnx/flexray-flux000-256-fp16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff01dfae72efcb5201bb4c24bcf22fc6a05b5fd619de8a033c475267073aba3b
+size 203297320

members/flux025/checksums.json

@@ -0,0 +1,9 @@
+{
+  "algorithm": "sha256",
+  "files": {
+    "config.yml": "7c7d52b2404d5e3cb5347036feac994ab8b8a18dd6bef5cfa47b749077af727a",
+    "label_schema.json": "6f60e2369859b7e17e275c04394c9ec33ac194a8118111577886b1c7fd3cf9f5",
+    "model.safetensors": "ba738a88f2473f3df11080f2106e577c475a40c351fb66c1410da25424cbf9f4",
+    "preprocessing.json": "5994ede07dc359e7fd62215ed1e97b811360fccacb6327bd91b6def439926155"
+  }
+}

members/flux025/config.yml

@@ -0,0 +1,87 @@
+model:
+  _class: fxr.models.UNet
+  in_channels: 1
+  out_channels: 61
+  filters:
+  - 64
+  - 128
+  - 256
+  - 512
+  - 512
+  - 720
+  - 1024
+  convs_per_block: 3
+  bottleneck_convs: null
+  skip_connections: true
+  dropout: 0.0
+  norm: instance
+  residual: true
+  norm_after_activation: true
+  upsample_align_corners: true
+  residual_projection: auto
+  residual_shortcut_norm: true
+protocol:
+  name: all_structures_flexray_v4
+  model_labels:
+    names:
+    - background
+    - skull
+    - scapulae
+    - clavicles
+    - humeri
+    - phalanges
+    - metacarpals
+    - carpals
+    - radii
+    - ulnae
+    - tibiae
+    - femurs
+    - patellae
+    - fibulae
+    - metatarsals
+    - tarsals
+    - toes
+    - rib_1
+    - rib_2
+    - rib_3
+    - rib_4
+    - rib_5
+    - rib_6
+    - rib_7
+    - rib_8
+    - rib_9
+    - rib_10
+    - rib_11
+    - rib_12
+    - sternum
+    - vertebra_c1
+    - vertebra_c2
+    - vertebra_c3
+    - vertebra_c4
+    - vertebra_c5
+    - vertebra_c6
+    - vertebra_c7
+    - vertebra_t1
+    - vertebra_t2
+    - vertebra_t3
+    - vertebra_t4
+    - vertebra_t5
+    - vertebra_t6
+    - vertebra_t7
+    - vertebra_t8
+    - vertebra_t9
+    - vertebra_t10
+    - vertebra_t11
+    - vertebra_t12
+    - vertebra_l1
+    - vertebra_l2
+    - vertebra_l3
+    - vertebra_l4
+    - vertebra_l5
+    - hips
+    - sacrum
+    - kidneys
+    - liver
+    - spleen
+    - lungs
+    - heart

members/flux025/label_schema.json

@@ -0,0 +1,70 @@
+{
+  "label_names": [
+    "background",
+    "skull",
+    "scapulae",
+    "clavicles",
+    "humeri",
+    "phalanges",
+    "metacarpals",
+    "carpals",
+    "radii",
+    "ulnae",
+    "tibiae",
+    "femurs",
+    "patellae",
+    "fibulae",
+    "metatarsals",
+    "tarsals",
+    "toes",
+    "rib_1",
+    "rib_2",
+    "rib_3",
+    "rib_4",
+    "rib_5",
+    "rib_6",
+    "rib_7",
+    "rib_8",
+    "rib_9",
+    "rib_10",
+    "rib_11",
+    "rib_12",
+    "sternum",
+    "vertebra_c1",
+    "vertebra_c2",
+    "vertebra_c3",
+    "vertebra_c4",
+    "vertebra_c5",
+    "vertebra_c6",
+    "vertebra_c7",
+    "vertebra_t1",
+    "vertebra_t2",
+    "vertebra_t3",
+    "vertebra_t4",
+    "vertebra_t5",
+    "vertebra_t6",
+    "vertebra_t7",
+    "vertebra_t8",
+    "vertebra_t9",
+    "vertebra_t10",
+    "vertebra_t11",
+    "vertebra_t12",
+    "vertebra_l1",
+    "vertebra_l2",
+    "vertebra_l3",
+    "vertebra_l4",
+    "vertebra_l5",
+    "hips",
+    "sacrum",
+    "kidneys",
+    "liver",
+    "spleen",
+    "lungs",
+    "heart"
+  ],
+  "label_order_verified": true,
+  "model_id": "VictorButoi/flexray",
+  "num_labels": 61,
+  "schema_version": 1,
+  "weights_license": "CC-BY-NC-4.0"
+}

members/flux025/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba738a88f2473f3df11080f2106e577c475a40c351fb66c1410da25424cbf9f4
+size 405630476

members/flux025/onnx/flexray-flux025-256-fp16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bde735f1514fa252c3179cbb21bdad1f442eefc1cd8d53aa34845b49e27186f5
+size 203297320

members/flux025/preprocessing.json

@@ -0,0 +1,20 @@
+{
+  "color_mode": "grayscale",
+  "image_size": [
+    256,
+    256
+  ],
+  "mean": null,
+  "normalization": {
+    "eps": 1e-08,
+    "percentiles": [
+      0.5,
+      99.5
+    ],
+    "scheme": "percentile_minmax"
+  },
+  "pad_to_square": true,
+  "probability_mode": "multilabel",
+  "scale": "zero_one",
+  "std": null
+}

checksums.json → members/flux0375/checksums.json

@@ -1,9 +1,8 @@
 {
   "algorithm": "sha256",
   "files": {
-    "README.md": "ce0a7a4087fec705e9cfe753bb01a06e0e2b29377e896e31e012f361e97b7058",
     "config.yml": "7c7d52b2404d5e3cb5347036feac994ab8b8a18dd6bef5cfa47b749077af727a",
-    "label_schema.json": "c8fa61b6585fc4043e2c51ba1f53271f9313cdae8d812479449b8a2a0ba12d7a",
+    "label_schema.json": "6f60e2369859b7e17e275c04394c9ec33ac194a8118111577886b1c7fd3cf9f5",
     "model.safetensors": "716ce3a8fb2a6abe22b2f2188fd596ddc4642b273fb0d336d629448154fc6d24",
     "preprocessing.json": "5994ede07dc359e7fd62215ed1e97b811360fccacb6327bd91b6def439926155"
   }

members/flux0375/config.yml

@@ -0,0 +1,87 @@
+model:
+  _class: fxr.models.UNet
+  in_channels: 1
+  out_channels: 61
+  filters:
+  - 64
+  - 128
+  - 256
+  - 512
+  - 512
+  - 720
+  - 1024
+  convs_per_block: 3
+  bottleneck_convs: null
+  skip_connections: true
+  dropout: 0.0
+  norm: instance
+  residual: true
+  norm_after_activation: true
+  upsample_align_corners: true
+  residual_projection: auto
+  residual_shortcut_norm: true
+protocol:
+  name: all_structures_flexray_v4
+  model_labels:
+    names:
+    - background
+    - skull
+    - scapulae
+    - clavicles
+    - humeri
+    - phalanges
+    - metacarpals
+    - carpals
+    - radii
+    - ulnae
+    - tibiae
+    - femurs
+    - patellae
+    - fibulae
+    - metatarsals
+    - tarsals
+    - toes
+    - rib_1
+    - rib_2
+    - rib_3
+    - rib_4
+    - rib_5
+    - rib_6
+    - rib_7
+    - rib_8
+    - rib_9
+    - rib_10
+    - rib_11
+    - rib_12
+    - sternum
+    - vertebra_c1
+    - vertebra_c2
+    - vertebra_c3
+    - vertebra_c4
+    - vertebra_c5
+    - vertebra_c6
+    - vertebra_c7
+    - vertebra_t1
+    - vertebra_t2
+    - vertebra_t3
+    - vertebra_t4
+    - vertebra_t5
+    - vertebra_t6
+    - vertebra_t7
+    - vertebra_t8
+    - vertebra_t9
+    - vertebra_t10
+    - vertebra_t11
+    - vertebra_t12
+    - vertebra_l1
+    - vertebra_l2
+    - vertebra_l3
+    - vertebra_l4
+    - vertebra_l5
+    - hips
+    - sacrum
+    - kidneys
+    - liver
+    - spleen
+    - lungs
+    - heart

members/flux0375/label_schema.json

@@ -0,0 +1,70 @@
+{
+  "label_names": [
+    "background",
+    "skull",
+    "scapulae",
+    "clavicles",
+    "humeri",
+    "phalanges",
+    "metacarpals",
+    "carpals",
+    "radii",
+    "ulnae",
+    "tibiae",
+    "femurs",
+    "patellae",
+    "fibulae",
+    "metatarsals",
+    "tarsals",
+    "toes",
+    "rib_1",
+    "rib_2",
+    "rib_3",
+    "rib_4",
+    "rib_5",
+    "rib_6",
+    "rib_7",
+    "rib_8",
+    "rib_9",
+    "rib_10",
+    "rib_11",
+    "rib_12",
+    "sternum",
+    "vertebra_c1",
+    "vertebra_c2",
+    "vertebra_c3",
+    "vertebra_c4",
+    "vertebra_c5",
+    "vertebra_c6",
+    "vertebra_c7",
+    "vertebra_t1",
+    "vertebra_t2",
+    "vertebra_t3",
+    "vertebra_t4",
+    "vertebra_t5",
+    "vertebra_t6",
+    "vertebra_t7",
+    "vertebra_t8",
+    "vertebra_t9",
+    "vertebra_t10",
+    "vertebra_t11",
+    "vertebra_t12",
+    "vertebra_l1",
+    "vertebra_l2",
+    "vertebra_l3",
+    "vertebra_l4",
+    "vertebra_l5",
+    "hips",
+    "sacrum",
+    "kidneys",
+    "liver",
+    "spleen",
+    "lungs",
+    "heart"
+  ],
+  "label_order_verified": true,
+  "model_id": "VictorButoi/flexray",
+  "num_labels": 61,
+  "schema_version": 1,
+  "weights_license": "CC-BY-NC-4.0"
+}

members/flux0375/preprocessing.json

@@ -0,0 +1,20 @@
+{
+  "color_mode": "grayscale",
+  "image_size": [
+    256,
+    256
+  ],
+  "mean": null,
+  "normalization": {
+    "eps": 1e-08,
+    "percentiles": [
+      0.5,
+      99.5
+    ],
+    "scheme": "percentile_minmax"
+  },
+  "pad_to_square": true,
+  "probability_mode": "multilabel",
+  "scale": "zero_one",
+  "std": null
+}

members/flux050/checksums.json

@@ -0,0 +1,9 @@
+{
+  "algorithm": "sha256",
+  "files": {
+    "config.yml": "7c7d52b2404d5e3cb5347036feac994ab8b8a18dd6bef5cfa47b749077af727a",
+    "label_schema.json": "6f60e2369859b7e17e275c04394c9ec33ac194a8118111577886b1c7fd3cf9f5",
+    "model.safetensors": "ebf580d17ad55c6316698e293837a11540c98dee07583812fe77690affed3edd",
+    "preprocessing.json": "5994ede07dc359e7fd62215ed1e97b811360fccacb6327bd91b6def439926155"
+  }
+}

members/flux050/config.yml

@@ -0,0 +1,87 @@
+model:
+  _class: fxr.models.UNet
+  in_channels: 1
+  out_channels: 61
+  filters:
+  - 64
+  - 128
+  - 256
+  - 512
+  - 512
+  - 720
+  - 1024
+  convs_per_block: 3
+  bottleneck_convs: null
+  skip_connections: true
+  dropout: 0.0
+  norm: instance
+  residual: true
+  norm_after_activation: true
+  upsample_align_corners: true
+  residual_projection: auto
+  residual_shortcut_norm: true
+protocol:
+  name: all_structures_flexray_v4
+  model_labels:
+    names:
+    - background
+    - skull
+    - scapulae
+    - clavicles
+    - humeri
+    - phalanges
+    - metacarpals
+    - carpals
+    - radii
+    - ulnae
+    - tibiae
+    - femurs
+    - patellae
+    - fibulae
+    - metatarsals
+    - tarsals
+    - toes
+    - rib_1
+    - rib_2
+    - rib_3
+    - rib_4
+    - rib_5
+    - rib_6
+    - rib_7
+    - rib_8
+    - rib_9
+    - rib_10
+    - rib_11
+    - rib_12
+    - sternum
+    - vertebra_c1
+    - vertebra_c2
+    - vertebra_c3
+    - vertebra_c4
+    - vertebra_c5
+    - vertebra_c6
+    - vertebra_c7
+    - vertebra_t1
+    - vertebra_t2
+    - vertebra_t3
+    - vertebra_t4
+    - vertebra_t5
+    - vertebra_t6
+    - vertebra_t7
+    - vertebra_t8
+    - vertebra_t9
+    - vertebra_t10
+    - vertebra_t11
+    - vertebra_t12
+    - vertebra_l1
+    - vertebra_l2
+    - vertebra_l3
+    - vertebra_l4
+    - vertebra_l5
+    - hips
+    - sacrum
+    - kidneys
+    - liver
+    - spleen
+    - lungs
+    - heart

members/flux050/label_schema.json

@@ -0,0 +1,70 @@
+{
+  "label_names": [
+    "background",
+    "skull",
+    "scapulae",
+    "clavicles",
+    "humeri",
+    "phalanges",
+    "metacarpals",
+    "carpals",
+    "radii",
+    "ulnae",
+    "tibiae",
+    "femurs",
+    "patellae",
+    "fibulae",
+    "metatarsals",
+    "tarsals",
+    "toes",
+    "rib_1",
+    "rib_2",
+    "rib_3",
+    "rib_4",
+    "rib_5",
+    "rib_6",
+    "rib_7",
+    "rib_8",
+    "rib_9",
+    "rib_10",
+    "rib_11",
+    "rib_12",
+    "sternum",
+    "vertebra_c1",
+    "vertebra_c2",
+    "vertebra_c3",
+    "vertebra_c4",
+    "vertebra_c5",
+    "vertebra_c6",
+    "vertebra_c7",
+    "vertebra_t1",
+    "vertebra_t2",
+    "vertebra_t3",
+    "vertebra_t4",
+    "vertebra_t5",
+    "vertebra_t6",
+    "vertebra_t7",
+    "vertebra_t8",
+    "vertebra_t9",
+    "vertebra_t10",
+    "vertebra_t11",
+    "vertebra_t12",
+    "vertebra_l1",
+    "vertebra_l2",
+    "vertebra_l3",
+    "vertebra_l4",
+    "vertebra_l5",
+    "hips",
+    "sacrum",
+    "kidneys",
+    "liver",
+    "spleen",
+    "lungs",
+    "heart"
+  ],
+  "label_order_verified": true,
+  "model_id": "VictorButoi/flexray",
+  "num_labels": 61,
+  "schema_version": 1,
+  "weights_license": "CC-BY-NC-4.0"
+}

members/flux050/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebf580d17ad55c6316698e293837a11540c98dee07583812fe77690affed3edd
+size 405630476

members/flux050/onnx/flexray-flux050-256-fp16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b729e19adf45956f081d3ac6093ea5f64147289b4ae2bb4e52ae07aafc3b8505
+size 203297320

members/flux050/preprocessing.json

@@ -0,0 +1,20 @@
+{
+  "color_mode": "grayscale",
+  "image_size": [
+    256,
+    256
+  ],
+  "mean": null,
+  "normalization": {
+    "eps": 1e-08,
+    "percentiles": [
+      0.5,
+      99.5
+    ],
+    "scheme": "percentile_minmax"
+  },
+  "pad_to_square": true,
+  "probability_mode": "multilabel",
+  "scale": "zero_one",
+  "std": null
+}

members/flux075/checksums.json

@@ -0,0 +1,9 @@
+{
+  "algorithm": "sha256",
+  "files": {
+    "config.yml": "7c7d52b2404d5e3cb5347036feac994ab8b8a18dd6bef5cfa47b749077af727a",
+    "label_schema.json": "6f60e2369859b7e17e275c04394c9ec33ac194a8118111577886b1c7fd3cf9f5",
+    "model.safetensors": "177a0bbc27d350c6a47de355b075a113882f5fb3fcd6d0e1314d7adf54821a5c",
+    "preprocessing.json": "5994ede07dc359e7fd62215ed1e97b811360fccacb6327bd91b6def439926155"
+  }
+}

members/flux075/config.yml

@@ -0,0 +1,87 @@
+model:
+  _class: fxr.models.UNet
+  in_channels: 1
+  out_channels: 61
+  filters:
+  - 64
+  - 128
+  - 256
+  - 512
+  - 512
+  - 720
+  - 1024
+  convs_per_block: 3
+  bottleneck_convs: null
+  skip_connections: true
+  dropout: 0.0
+  norm: instance
+  residual: true
+  norm_after_activation: true
+  upsample_align_corners: true
+  residual_projection: auto
+  residual_shortcut_norm: true
+protocol:
+  name: all_structures_flexray_v4
+  model_labels:
+    names:
+    - background
+    - skull
+    - scapulae
+    - clavicles
+    - humeri
+    - phalanges
+    - metacarpals
+    - carpals
+    - radii
+    - ulnae
+    - tibiae
+    - femurs
+    - patellae
+    - fibulae
+    - metatarsals
+    - tarsals
+    - toes
+    - rib_1
+    - rib_2
+    - rib_3
+    - rib_4
+    - rib_5
+    - rib_6
+    - rib_7
+    - rib_8
+    - rib_9
+    - rib_10
+    - rib_11
+    - rib_12
+    - sternum
+    - vertebra_c1
+    - vertebra_c2
+    - vertebra_c3
+    - vertebra_c4
+    - vertebra_c5
+    - vertebra_c6
+    - vertebra_c7
+    - vertebra_t1
+    - vertebra_t2
+    - vertebra_t3
+    - vertebra_t4
+    - vertebra_t5
+    - vertebra_t6
+    - vertebra_t7
+    - vertebra_t8
+    - vertebra_t9
+    - vertebra_t10
+    - vertebra_t11
+    - vertebra_t12
+    - vertebra_l1
+    - vertebra_l2
+    - vertebra_l3
+    - vertebra_l4
+    - vertebra_l5
+    - hips
+    - sacrum
+    - kidneys
+    - liver
+    - spleen
+    - lungs
+    - heart

members/flux075/label_schema.json

@@ -0,0 +1,70 @@
+{
+  "label_names": [
+    "background",
+    "skull",
+    "scapulae",
+    "clavicles",
+    "humeri",
+    "phalanges",
+    "metacarpals",
+    "carpals",
+    "radii",
+    "ulnae",
+    "tibiae",
+    "femurs",
+    "patellae",
+    "fibulae",
+    "metatarsals",
+    "tarsals",
+    "toes",
+    "rib_1",
+    "rib_2",
+    "rib_3",
+    "rib_4",
+    "rib_5",
+    "rib_6",
+    "rib_7",
+    "rib_8",
+    "rib_9",
+    "rib_10",
+    "rib_11",
+    "rib_12",
+    "sternum",
+    "vertebra_c1",
+    "vertebra_c2",
+    "vertebra_c3",
+    "vertebra_c4",
+    "vertebra_c5",
+    "vertebra_c6",
+    "vertebra_c7",
+    "vertebra_t1",
+    "vertebra_t2",
+    "vertebra_t3",
+    "vertebra_t4",
+    "vertebra_t5",
+    "vertebra_t6",
+    "vertebra_t7",
+    "vertebra_t8",
+    "vertebra_t9",
+    "vertebra_t10",
+    "vertebra_t11",
+    "vertebra_t12",
+    "vertebra_l1",
+    "vertebra_l2",
+    "vertebra_l3",
+    "vertebra_l4",
+    "vertebra_l5",
+    "hips",
+    "sacrum",
+    "kidneys",
+    "liver",
+    "spleen",
+    "lungs",
+    "heart"
+  ],
+  "label_order_verified": true,
+  "model_id": "VictorButoi/flexray",
+  "num_labels": 61,
+  "schema_version": 1,
+  "weights_license": "CC-BY-NC-4.0"
+}

members/flux075/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:177a0bbc27d350c6a47de355b075a113882f5fb3fcd6d0e1314d7adf54821a5c
+size 405630476

members/flux075/onnx/flexray-flux075-256-fp16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d94b18c53e5d47c65aaeede0155f9c4b557df1a2e7fd926b43dc992e21521c63
+size 203297320

members/flux075/preprocessing.json

@@ -0,0 +1,20 @@
+{
+  "color_mode": "grayscale",
+  "image_size": [
+    256,
+    256
+  ],
+  "mean": null,
+  "normalization": {
+    "eps": 1e-08,
+    "percentiles": [
+      0.5,
+      99.5
+    ],
+    "scheme": "percentile_minmax"
+  },
+  "pad_to_square": true,
+  "probability_mode": "multilabel",
+  "scale": "zero_one",
+  "std": null
+}