Add model

README.md

@@ -0,0 +1,12 @@
+---
+tags:
+- image-feature-extraction
+- timm
+library_name: timm
+license: other
+license_name: apple-sample-code-license
+license_link: https://huggingface.co/apple/DFN5B-CLIP-ViT-H-14-378/resolve/main/LICENSE
+---
+# Model card for vit_huge_patch14_clip_378.dfn5b
+
+`timm` DFN CLIP (image encoder only) weights from https://huggingface.co/apple/DFN5B-CLIP-ViT-H-14-378

config.json

@@ -0,0 +1,36 @@
+{
+  "architecture": "vit_huge_patch14_clip_378",
+  "num_classes": 1024,
+  "num_features": 1280,
+  "global_pool": "token",
+  "pretrained_cfg": {
+    "tag": "dfn5b",
+    "custom_load": false,
+    "input_size": [
+      3,
+      378,
+      378
+    ],
+    "fixed_input_size": true,
+    "interpolation": "bicubic",
+    "crop_pct": 1.0,
+    "crop_mode": "center",
+    "mean": [
+      0.48145466,
+      0.4578275,
+      0.40821073
+    ],
+    "std": [
+      0.26862954,
+      0.26130258,
+      0.27577711
+    ],
+    "num_classes": 1024,
+    "pool_size": null,
+    "first_conv": "patch_embed.proj",
+    "classifier": "head",
+    "notes": [
+      "natively QuickGELU, use quickgelu model variant for original results"
+    ]
+  }
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81d5bf47b916583e0e1ad4d74863823a8eac90de7752ef921fc9e14df3c78b30
+size 2530770824

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc0d552f1bcb2e0d11603efcf8f2f6619f8200001198cb1bc21b713487312458
+size 2530881006