Instructions to use kerasformers/sam3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- KerasFormers
How to use kerasformers/sam3 with KerasFormers:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Keras
How to use kerasformers/sam3 with Keras:
# Available backend options are: "jax", "torch", "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" import keras model = keras.saving.load_model("hf://kerasformers/sam3") - Notebooks
- Google Colab
- Kaggle
| { | |
| "library_name": "kerasformers", | |
| "kerasformers_version": "1.1.3", | |
| "model_module": "kerasformers.models.sam3", | |
| "model_class": "SAM3Model", | |
| "variant": "sam3_saco", | |
| "weights": "model.weights.h5", | |
| "model_type": "sam3", | |
| "vit_hidden_size": 1024, | |
| "vit_intermediate_size": 4736, | |
| "vit_num_hidden_layers": 32, | |
| "vit_num_attention_heads": 16, | |
| "vit_image_size": 1008, | |
| "vit_patch_size": 14, | |
| "vit_window_size": 24, | |
| "vit_global_attn_indexes": [ | |
| 7, | |
| 15, | |
| 23, | |
| 31 | |
| ], | |
| "vit_rope_theta": 10000.0, | |
| "vit_pretrain_image_size": 336, | |
| "fpn_hidden_size": 256, | |
| "fpn_scale_factors": [ | |
| 4.0, | |
| 2.0, | |
| 1.0, | |
| 0.5 | |
| ], | |
| "detr_encoder_hidden_size": 256, | |
| "detr_encoder_num_layers": 6, | |
| "detr_encoder_num_attention_heads": 8, | |
| "detr_encoder_intermediate_size": 2048, | |
| "detr_encoder_dropout": 0.1, | |
| "detr_decoder_hidden_size": 256, | |
| "detr_decoder_num_layers": 6, | |
| "detr_decoder_num_queries": 200, | |
| "detr_decoder_num_attention_heads": 8, | |
| "detr_decoder_intermediate_size": 2048, | |
| "detr_decoder_dropout": 0.1, | |
| "mask_decoder_hidden_size": 256, | |
| "mask_decoder_num_upsampling_stages": 3, | |
| "mask_decoder_num_attention_heads": 8, | |
| "text_hidden_size": 1024, | |
| "text_projection_dim": 512, | |
| "image_size": 1008 | |
| } |