Update tiny models for DepthProForDepthEstimation

config.json

@@ -0,0 +1,135 @@
+{
+  "architectures": [
+    "DepthProForDepthEstimation"
+  ],
+  "dtype": "float32",
+  "fov_model_config": {
+    "apply_layernorm": true,
+    "attention_probs_dropout_prob": 0.0,
+    "drop_path_rate": 0.0,
+    "hidden_act": "gelu",
+    "hidden_dropout_prob": 0.0,
+    "hidden_size": 32,
+    "image_size": 16,
+    "initializer_range": 0.02,
+    "layer_norm_eps": 1e-06,
+    "layerscale_value": 1.0,
+    "mlp_ratio": 4,
+    "model_type": "dinov2",
+    "num_attention_heads": 4,
+    "num_channels": 3,
+    "num_hidden_layers": 2,
+    "out_features": [
+      "stage2"
+    ],
+    "out_indices": [
+      2
+    ],
+    "patch_size": 8,
+    "qkv_bias": true,
+    "reshape_hidden_states": true,
+    "stage_names": [
+      "stem",
+      "stage1",
+      "stage2"
+    ],
+    "use_mask_token": true,
+    "use_swiglu_ffn": false
+  },
+  "fusion_hidden_size": 16,
+  "image_model_config": {
+    "apply_layernorm": true,
+    "attention_probs_dropout_prob": 0.0,
+    "drop_path_rate": 0.0,
+    "hidden_act": "gelu",
+    "hidden_dropout_prob": 0.0,
+    "hidden_size": 16,
+    "image_size": 16,
+    "initializer_range": 0.02,
+    "layer_norm_eps": 1e-06,
+    "layerscale_value": 1.0,
+    "mlp_ratio": 4,
+    "model_type": "dinov2",
+    "num_attention_heads": 1,
+    "num_channels": 3,
+    "num_hidden_layers": 2,
+    "out_features": [
+      "stage2"
+    ],
+    "out_indices": [
+      2
+    ],
+    "patch_size": 4,
+    "qkv_bias": true,
+    "reshape_hidden_states": true,
+    "stage_names": [
+      "stem",
+      "stage1",
+      "stage2"
+    ],
+    "use_mask_token": true,
+    "use_swiglu_ffn": false
+  },
+  "initializer_range": 0.02,
+  "intermediate_feature_dims": [
+    10,
+    8
+  ],
+  "intermediate_hook_ids": [
+    1,
+    0
+  ],
+  "merge_padding_value": 3,
+  "model_type": "depth_pro",
+  "num_fov_head_layers": 2,
+  "patch_model_config": {
+    "apply_layernorm": true,
+    "attention_probs_dropout_prob": 0.0,
+    "drop_path_rate": 0.0,
+    "hidden_act": "gelu",
+    "hidden_dropout_prob": 0.0,
+    "hidden_size": 24,
+    "image_size": 16,
+    "initializer_range": 0.02,
+    "layer_norm_eps": 1e-06,
+    "layerscale_value": 1.0,
+    "mlp_ratio": 4,
+    "model_type": "dinov2",
+    "num_attention_heads": 2,
+    "num_channels": 3,
+    "num_hidden_layers": 2,
+    "out_features": [
+      "stage2"
+    ],
+    "out_indices": [
+      2
+    ],
+    "patch_size": 6,
+    "qkv_bias": true,
+    "reshape_hidden_states": true,
+    "stage_names": [
+      "stem",
+      "stage1",
+      "stage2"
+    ],
+    "use_mask_token": true,
+    "use_swiglu_ffn": false
+  },
+  "patch_size": 16,
+  "scaled_images_feature_dims": [
+    12,
+    12
+  ],
+  "scaled_images_overlap_ratios": [
+    0.0,
+    0.2
+  ],
+  "scaled_images_ratios": [
+    0.5,
+    1.0
+  ],
+  "transformers_version": "5.16.0.dev0",
+  "use_batch_norm_in_fusion_residual": false,
+  "use_bias_in_fusion_residual": true,
+  "use_fov_model": false
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b97ffc4bfb9da88232aaa27e529d1026258eedd1ac43973ae080a1252a9a0252
+size 345716

preprocessor_config.json

@@ -0,0 +1,22 @@
+{
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "DepthProImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 1536,
+    "width": 1536
+  }
+}