Update tiny models for ZoeDepthForDepthEstimation

config.json

@@ -0,0 +1,98 @@
+{
+  "add_projection": false,
+  "architectures": [
+    "ZoeDepthForDepthEstimation"
+  ],
+  "attractor_alpha": 1000,
+  "attractor_gamma": 2,
+  "attractor_kind": "mean",
+  "backbone": null,
+  "backbone_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": 4,
+    "image_size": 32,
+    "initializer_range": 0.02,
+    "intermediate_size": 8,
+    "is_training": true,
+    "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": [
+      "stage1",
+      "stage2"
+    ],
+    "out_indices": [
+      1,
+      2
+    ],
+    "patch_size": 16,
+    "qkv_bias": true,
+    "reshape_hidden_states": false,
+    "stage_names": [
+      "stem",
+      "stage1",
+      "stage2"
+    ],
+    "use_mask_token": true,
+    "use_swiglu_ffn": false
+  },
+  "backbone_hidden_size": 4,
+  "batch_norm_eps": 1e-05,
+  "bin_centers_type": "softplus",
+  "bin_configurations": [
+    {
+      "max_depth": 10.0,
+      "min_depth": 0.001,
+      "n_bins": 64
+    }
+  ],
+  "bin_embedding_dim": 128,
+  "bottleneck_features": 6,
+  "dtype": "float32",
+  "fusion_hidden_size": 6,
+  "head_in_index": -1,
+  "hidden_act": "gelu",
+  "initializer_range": 0.02,
+  "max_temp": 50.0,
+  "min_temp": 0.0212,
+  "model_type": "zoedepth",
+  "neck_hidden_sizes": [
+    2,
+    2
+  ],
+  "num_attractors": [
+    16,
+    8,
+    4,
+    1
+  ],
+  "num_out_features": [
+    6,
+    6,
+    6,
+    6
+  ],
+  "num_patch_transformer_layers": null,
+  "num_relative_features": 32,
+  "patch_transformer_hidden_size": null,
+  "patch_transformer_intermediate_size": null,
+  "patch_transformer_num_attention_heads": null,
+  "readout_type": "project",
+  "reassemble_factors": [
+    4,
+    2,
+    1,
+    0.5
+  ],
+  "transformers_version": "5.16.0.dev0",
+  "use_batch_norm_in_fusion_residual": false,
+  "use_bias_in_fusion_residual": null
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:399ec6a6980b24e538bfd87ae085ba300a63d7748ac5f3670fa3ef14eaa12f02
+size 798708

preprocessor_config.json

@@ -0,0 +1,25 @@
+{
+  "do_normalize": true,
+  "do_pad": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "ensure_multiple_of": 32,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "ZoeDepthImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "keep_aspect_ratio": true,
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 384,
+    "width": 512
+  }
+}