Add/Update backbone checkpoints (count=6)

README.md

@@ -1,11 +1,16 @@
-# Backbone+MLPHead checkpoints (single repo)
+---
+license: apache-2.0
+library_name: transformers
+pipeline_tag: image-classification
+tags:
+  - backbone
+  - checkpoint
+---
 
-- Repo: `dsaint31/bb_mlp_224`
+# Root README
 
-## Layout (구조)
-- `models/<backbone_sanitized>/` 아래에 backbone별 체크포인트가 있습니다.
+Repo: `dsaint31/bb_mlp_224`
 
-## Available checkpoints (목록)
 - `google/vit-base-patch16-224` -> `models/google__vit-base-patch16-224`
 - `microsoft/swin-tiny-patch4-window7-224` -> `models/microsoft__swin-tiny-patch4-window7-224`
 - `microsoft/resnet-50` -> `models/microsoft__resnet-50`
@@ -13,22 +18,3 @@
 - `timm/densenet121.tv_in1k` -> `models/timm__densenet121.tv_in1k`
 - `torchvision/densenet121` -> `models/torchvision__densenet121`
 
-## How to load (로드 예시)
-```python
-from huggingface_hub import snapshot_download
-from ds_model import BackboneWithMLPHeadForImageClassification
-
-repo_id = "dsaint31/bb_mlp_224"
-revision = "main"
-local = snapshot_download(repo_id=repo_id, revision=revision)
-
-# pick one subfolder
-subdir = "models/<backbone_sanitized>"
-model_path = f"{local}/{subdir}"
-
-model = BackboneWithMLPHeadForImageClassification.from_pretrained(model_path)
-# IMPORTANT: do NOT call model.load_backbone_pretrained_() here
-```
-
-## Notes (비고)
-- provenance는 각 subfolder의 `config.json` -> `ds_provenance`에 저장됩니다.

manifest_20260210_050905.json

@@ -0,0 +1,41 @@
+{
+  "timestamp": "20260210_050905",
+  "device": "mps",
+  "repo_id": "dsaint31/bb_mlp_224",
+  "revision": "main",
+  "tag": null,
+  "num_labels": 3,
+  "count": 6,
+  "items": [
+    {
+      "backbone": "google/vit-base-patch16-224",
+      "subdir": "models/google__vit-base-patch16-224",
+      "dirname": "google__vit-base-patch16-224"
+    },
+    {
+      "backbone": "microsoft/swin-tiny-patch4-window7-224",
+      "subdir": "models/microsoft__swin-tiny-patch4-window7-224",
+      "dirname": "microsoft__swin-tiny-patch4-window7-224"
+    },
+    {
+      "backbone": "microsoft/resnet-50",
+      "subdir": "models/microsoft__resnet-50",
+      "dirname": "microsoft__resnet-50"
+    },
+    {
+      "backbone": "google/efficientnet-b0",
+      "subdir": "models/google__efficientnet-b0",
+      "dirname": "google__efficientnet-b0"
+    },
+    {
+      "backbone": "timm/densenet121.tv_in1k",
+      "subdir": "models/timm__densenet121.tv_in1k",
+      "dirname": "timm__densenet121.tv_in1k"
+    },
+    {
+      "backbone": "torchvision/densenet121",
+      "subdir": "models/torchvision__densenet121",
+      "dirname": "torchvision__densenet121"
+    }
+  ]
+}

models/google__efficientnet-b0/README.md

@@ -1,28 +1,14 @@
-# Backbone+MLPHead checkpoint (per-backbone)
-
-- Repo: `dsaint31/bb_mlp_224`
-- Subdir: `models/google__efficientnet-b0`
-- Backbone: `google/efficientnet-b0`
-- num_labels: `3`
-- Revision: `main`
-
-## Load (로드)
-
-```python
-from huggingface_hub import snapshot_download
-from ds_model import BackboneWithMLPHeadForImageClassification
-
-repo_id = "dsaint31/bb_mlp_224"
-revision = "main"
-local = snapshot_download(repo_id=repo_id, revision=revision)
-
-model_path = f"{local}/models/google__efficientnet-b0"
-model = BackboneWithMLPHeadForImageClassification.from_pretrained(model_path)
-# IMPORTANT: do NOT call model.load_backbone_pretrained_() here
-```
-
-## Design policy (설계 원칙)
-
-- __init__()에서 pretrained backbone 로드 금지
-- fresh-start 체크포인트 생성 시에만 load_backbone_pretrained_()로 주입
-- from_pretrained 이후 load_backbone_pretrained_() 호출 금지
+---
+license: apache-2.0
+pipeline_tag: image-classification
+tags:
+  - backbone
+  - checkpoint
+  - subfolder
+---
+
+# Per-backbone README
+
+Backbone: `google/efficientnet-b0`
+Subdir: `models/google__efficientnet-b0`
+num_labels: `3`

models/google__efficientnet-b0/config.json

@@ -24,7 +24,7 @@
   "num_labels": 3,
   "transformers_version": "5.1.0",
   "ds_provenance": {
-    "created_at": "20260210_045210",
+    "created_at": "20260210_050905",
     "repo_id": "dsaint31/bb_mlp_224",
     "subdir": "models/google__efficientnet-b0",
     "wrapper_class": "BackboneWithMLPHeadForImageClassification",

models/google__efficientnet-b0/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:355d7175f8ab37ac6bfaa15e76cb9263d95071ed967eb646067d918df477e43e
+oid sha256:a67fc215d299f5fe53f61da21d454c40f414d1221ff57167797cc276ee5e2991
 size 17558436

models/google__vit-base-patch16-224/README.md

@@ -1,28 +1,14 @@
-# Backbone+MLPHead checkpoint (per-backbone)
-
-- Repo: `dsaint31/bb_mlp_224`
-- Subdir: `models/google__vit-base-patch16-224`
-- Backbone: `google/vit-base-patch16-224`
-- num_labels: `3`
-- Revision: `main`
-
-## Load (로드)
-
-```python
-from huggingface_hub import snapshot_download
-from ds_model import BackboneWithMLPHeadForImageClassification
-
-repo_id = "dsaint31/bb_mlp_224"
-revision = "main"
-local = snapshot_download(repo_id=repo_id, revision=revision)
-
-model_path = f"{local}/models/google__vit-base-patch16-224"
-model = BackboneWithMLPHeadForImageClassification.from_pretrained(model_path)
-# IMPORTANT: do NOT call model.load_backbone_pretrained_() here
-```
-
-## Design policy (설계 원칙)
-
-- __init__()에서 pretrained backbone 로드 금지
-- fresh-start 체크포인트 생성 시에만 load_backbone_pretrained_()로 주입
-- from_pretrained 이후 load_backbone_pretrained_() 호출 금지
+---
+license: apache-2.0
+pipeline_tag: image-classification
+tags:
+  - backbone
+  - checkpoint
+  - subfolder
+---
+
+# Per-backbone README
+
+Backbone: `google/vit-base-patch16-224`
+Subdir: `models/google__vit-base-patch16-224`
+num_labels: `3`

models/google__vit-base-patch16-224/config.json

@@ -24,7 +24,7 @@
   "num_labels": 3,
   "transformers_version": "5.1.0",
   "ds_provenance": {
-    "created_at": "20260210_045210",
+    "created_at": "20260210_050905",
     "repo_id": "dsaint31/bb_mlp_224",
     "subdir": "models/google__vit-base-patch16-224",
     "wrapper_class": "BackboneWithMLPHeadForImageClassification",

models/google__vit-base-patch16-224/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:966da5497febbf805e9c5cb706350f00809d56b02babe8ad3ad388631153b866
+oid sha256:3d8f30959f7f4875a6e7792d2e5d27ec06ea99c166fbe9bdb9aab09173fa784d
 size 346372132

models/microsoft__resnet-50/README.md

@@ -1,28 +1,14 @@
-# Backbone+MLPHead checkpoint (per-backbone)
-
-- Repo: `dsaint31/bb_mlp_224`
-- Subdir: `models/microsoft__resnet-50`
-- Backbone: `microsoft/resnet-50`
-- num_labels: `3`
-- Revision: `main`
-
-## Load (로드)
-
-```python
-from huggingface_hub import snapshot_download
-from ds_model import BackboneWithMLPHeadForImageClassification
-
-repo_id = "dsaint31/bb_mlp_224"
-revision = "main"
-local = snapshot_download(repo_id=repo_id, revision=revision)
-
-model_path = f"{local}/models/microsoft__resnet-50"
-model = BackboneWithMLPHeadForImageClassification.from_pretrained(model_path)
-# IMPORTANT: do NOT call model.load_backbone_pretrained_() here
-```
-
-## Design policy (설계 원칙)
-
-- __init__()에서 pretrained backbone 로드 금지
-- fresh-start 체크포인트 생성 시에만 load_backbone_pretrained_()로 주입
-- from_pretrained 이후 load_backbone_pretrained_() 호출 금지
+---
+license: apache-2.0
+pipeline_tag: image-classification
+tags:
+  - backbone
+  - checkpoint
+  - subfolder
+---
+
+# Per-backbone README
+
+Backbone: `microsoft/resnet-50`
+Subdir: `models/microsoft__resnet-50`
+num_labels: `3`

models/microsoft__resnet-50/config.json

@@ -24,7 +24,7 @@
   "num_labels": 3,
   "transformers_version": "5.1.0",
   "ds_provenance": {
-    "created_at": "20260210_045210",
+    "created_at": "20260210_050905",
     "repo_id": "dsaint31/bb_mlp_224",
     "subdir": "models/microsoft__resnet-50",
     "wrapper_class": "BackboneWithMLPHeadForImageClassification",

models/microsoft__resnet-50/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b7241b91469e351ec6e9b961eb614c503bbd6d761993242a50ce47deca5de202
+oid sha256:12f287e717506b8e251dddbc02e8471a753795da5c1ea503c3491817f61f8c63
 size 96388660

models/microsoft__swin-tiny-patch4-window7-224/README.md

@@ -1,28 +1,14 @@
-# Backbone+MLPHead checkpoint (per-backbone)
-
-- Repo: `dsaint31/bb_mlp_224`
-- Subdir: `models/microsoft__swin-tiny-patch4-window7-224`
-- Backbone: `microsoft/swin-tiny-patch4-window7-224`
-- num_labels: `3`
-- Revision: `main`
-
-## Load (로드)
-
-```python
-from huggingface_hub import snapshot_download
-from ds_model import BackboneWithMLPHeadForImageClassification
-
-repo_id = "dsaint31/bb_mlp_224"
-revision = "main"
-local = snapshot_download(repo_id=repo_id, revision=revision)
-
-model_path = f"{local}/models/microsoft__swin-tiny-patch4-window7-224"
-model = BackboneWithMLPHeadForImageClassification.from_pretrained(model_path)
-# IMPORTANT: do NOT call model.load_backbone_pretrained_() here
-```
-
-## Design policy (설계 원칙)
-
-- __init__()에서 pretrained backbone 로드 금지
-- fresh-start 체크포인트 생성 시에만 load_backbone_pretrained_()로 주입
-- from_pretrained 이후 load_backbone_pretrained_() 호출 금지
+---
+license: apache-2.0
+pipeline_tag: image-classification
+tags:
+  - backbone
+  - checkpoint
+  - subfolder
+---
+
+# Per-backbone README
+
+Backbone: `microsoft/swin-tiny-patch4-window7-224`
+Subdir: `models/microsoft__swin-tiny-patch4-window7-224`
+num_labels: `3`

models/microsoft__swin-tiny-patch4-window7-224/config.json

@@ -24,7 +24,7 @@
   "num_labels": 3,
   "transformers_version": "5.1.0",
   "ds_provenance": {
-    "created_at": "20260210_045210",
+    "created_at": "20260210_050905",
     "repo_id": "dsaint31/bb_mlp_224",
     "subdir": "models/microsoft__swin-tiny-patch4-window7-224",
     "wrapper_class": "BackboneWithMLPHeadForImageClassification",

models/microsoft__swin-tiny-patch4-window7-224/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f05be6029d1cd0bb50a44295ce9b4d4590d25db5004768cb0bafd2217a00f105
+oid sha256:045a8d82d86041b915ef92e1bfc62b1ff3dd85e034f87025c9e6e68d018c40ba
 size 111128348

models/timm__densenet121.tv_in1k/README.md

@@ -1,28 +1,14 @@
-# Backbone+MLPHead checkpoint (per-backbone)
-
-- Repo: `dsaint31/bb_mlp_224`
-- Subdir: `models/timm__densenet121.tv_in1k`
-- Backbone: `timm/densenet121.tv_in1k`
-- num_labels: `3`
-- Revision: `main`
-
-## Load (로드)
-
-```python
-from huggingface_hub import snapshot_download
-from ds_model import BackboneWithMLPHeadForImageClassification
-
-repo_id = "dsaint31/bb_mlp_224"
-revision = "main"
-local = snapshot_download(repo_id=repo_id, revision=revision)
-
-model_path = f"{local}/models/timm__densenet121.tv_in1k"
-model = BackboneWithMLPHeadForImageClassification.from_pretrained(model_path)
-# IMPORTANT: do NOT call model.load_backbone_pretrained_() here
-```
-
-## Design policy (설계 원칙)
-
-- __init__()에서 pretrained backbone 로드 금지
-- fresh-start 체크포인트 생성 시에만 load_backbone_pretrained_()로 주입
-- from_pretrained 이후 load_backbone_pretrained_() 호출 금지
+---
+license: apache-2.0
+pipeline_tag: image-classification
+tags:
+  - backbone
+  - checkpoint
+  - subfolder
+---
+
+# Per-backbone README
+
+Backbone: `timm/densenet121.tv_in1k`
+Subdir: `models/timm__densenet121.tv_in1k`
+num_labels: `3`

models/timm__densenet121.tv_in1k/config.json

@@ -24,7 +24,7 @@
   "num_labels": 3,
   "transformers_version": "5.1.0",
   "ds_provenance": {
-    "created_at": "20260210_045210",
+    "created_at": "20260210_050905",
     "repo_id": "dsaint31/bb_mlp_224",
     "subdir": "models/timm__densenet121.tv_in1k",
     "wrapper_class": "BackboneWithMLPHeadForImageClassification",

models/timm__densenet121.tv_in1k/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d4feb04b67f845127dd33382323bbfe50bf9b31c7060d2bb6a700b7ac55401c
+oid sha256:5c7a629e31d700969130554fdfe93c8cd2f9344fb865b6903c3df151c0624ec6
 size 29293620

models/torchvision__densenet121/README.md

@@ -1,28 +1,14 @@
-# Backbone+MLPHead checkpoint (per-backbone)
-
-- Repo: `dsaint31/bb_mlp_224`
-- Subdir: `models/torchvision__densenet121`
-- Backbone: `torchvision/densenet121`
-- num_labels: `3`
-- Revision: `main`
-
-## Load (로드)
-
-```python
-from huggingface_hub import snapshot_download
-from ds_model import BackboneWithMLPHeadForImageClassification
-
-repo_id = "dsaint31/bb_mlp_224"
-revision = "main"
-local = snapshot_download(repo_id=repo_id, revision=revision)
-
-model_path = f"{local}/models/torchvision__densenet121"
-model = BackboneWithMLPHeadForImageClassification.from_pretrained(model_path)
-# IMPORTANT: do NOT call model.load_backbone_pretrained_() here
-```
-
-## Design policy (설계 원칙)
-
-- __init__()에서 pretrained backbone 로드 금지
-- fresh-start 체크포인트 생성 시에만 load_backbone_pretrained_()로 주입
-- from_pretrained 이후 load_backbone_pretrained_() 호출 금지
+---
+license: apache-2.0
+pipeline_tag: image-classification
+tags:
+  - backbone
+  - checkpoint
+  - subfolder
+---
+
+# Per-backbone README
+
+Backbone: `torchvision/densenet121`
+Subdir: `models/torchvision__densenet121`
+num_labels: `3`

models/torchvision__densenet121/config.json

@@ -24,7 +24,7 @@
   "num_labels": 3,
   "transformers_version": "5.1.0",
   "ds_provenance": {
-    "created_at": "20260210_045210",
+    "created_at": "20260210_050905",
     "repo_id": "dsaint31/bb_mlp_224",
     "subdir": "models/torchvision__densenet121",
     "wrapper_class": "BackboneWithMLPHeadForImageClassification",

models/torchvision__densenet121/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5069db9d5d1992568c493e7b2c4d153307bbdb55ad5c53a8ae99c4800e818aff
+oid sha256:969ad71b68ca03294c98303e609ea596d833f4381ff3039131e4a5239b72dd99
 size 33394052