Image Feature Extraction
Transformers
Safetensors
page
feature-extraction
gaze-estimation
gaze-target-estimation
dinov3
custom_code
Instructions to use Octopus1/page-vits with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Octopus1/page-vits with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-feature-extraction", model="Octopus1/page-vits", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Octopus1/page-vits", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
File size: 1,157 Bytes
7c1f9ab aa8ddb4 7c1f9ab aa8ddb4 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 | {
"architectures": [
"PaGEModel"
],
"dim": 256,
"dino_feature_dropout": 0.1,
"dinov3_drop_path_rate": 0.0,
"dinov3_hidden_size": 384,
"dinov3_intermediate_size": 1536,
"dinov3_layer_norm_eps": 1e-05,
"dinov3_layerscale_value": 1.0,
"dinov3_num_attention_heads": 6,
"dinov3_num_hidden_layers": 12,
"dinov3_num_register_tokens": 4,
"dinov3_patch_size": 16,
"dinov3_use_gated_mlp": false,
"drop_path": 0.1,
"dtype": "float32",
"head_in_size": [
256,
256
],
"heatmap_out_size": [
64,
64
],
"image_mean": [
0.485,
0.456,
0.406
],
"image_std": [
0.229,
0.224,
0.225
],
"inout": true,
"mlp_layer": "geglu",
"mlp_ratio": 4.0,
"model_type": "page",
"n_head_self_attn_layers": 1,
"n_reg_tokens": 4,
"n_scene_head_interaction_layers": 5,
"n_scene_self_attn_layers": 1,
"num_heads": 8,
"pos_encoding": "rope",
"scene_in_size": [
512,
512
],
"transformers_version": "5.6.2",
"use_head_prompt": false,
"auto_map": {
"AutoConfig": "Octopus1/PaGE--modeling_page.PaGEConfig",
"AutoModel": "Octopus1/PaGE--modeling_page.PaGEModel"
}
}
|