|
|
--- |
|
|
library_name: keras-hub |
|
|
--- |
|
|
This is a [`DepthAnything` model](https://keras.io/api/keras_hub/models/depth_anything) uploaded using the KerasHub library and can be used with JAX, TensorFlow, and PyTorch backends. |
|
|
Model config: |
|
|
* **name:** depth_anything_backbone |
|
|
* **trainable:** True |
|
|
* **dtype:** {'module': 'keras', 'class_name': 'DTypePolicy', 'config': {'name': 'float32'}, 'registered_name': None} |
|
|
* **image_encoder:** {'module': 'keras_hub.src.models.dinov2.dinov2_backbone', 'class_name': 'DINOV2Backbone', 'config': {'name': 'dinov2_backbone', 'trainable': True, 'dtype': {'module': 'keras', 'class_name': 'DTypePolicy', 'config': {'name': 'float32'}, 'registered_name': None}, 'patch_size': 14, 'num_layers': 12, 'hidden_dim': 768, 'num_heads': 12, 'intermediate_dim': 3072, 'layer_scale_init_value': 1.0, 'num_register_tokens': 0, 'use_mask_token': True, 'use_swiglu_ffn': False, 'dropout_rate': 0.0, 'drop_path_rate': 0.0, 'image_shape': [518, 518, 3], 'position_embedding_shape': [518, 518], 'antialias_in_interpolation': False, 'apply_layernorm': True}, 'registered_name': 'keras_hub>DINOV2Backbone'} |
|
|
* **reassemble_factors:** [4, 2, 1, 0.5] |
|
|
* **neck_hidden_dims:** [96, 192, 384, 768] |
|
|
* **fusion_hidden_dim:** 128 |
|
|
* **head_hidden_dim:** 32 |
|
|
* **head_in_index:** -1 |
|
|
* **feature_keys:** ['stage3', 'stage6', 'stage9', 'stage12'] |
|
|
|
|
|
This model card has been generated automatically and should be completed by the model author. See [Model Cards documentation](https://huggingface.co/docs/hub/model-cards) for more information. |
|
|
|
