Instructions to use james-emi/darpa_ir-f8v2-person-mask2former-swinb-v4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use james-emi/darpa_ir-f8v2-person-mask2former-swinb-v4 with Transformers:
# Load model directly from transformers import AutoImageProcessor, Mask2FormerForUniversalSegmentation processor = AutoImageProcessor.from_pretrained("james-emi/darpa_ir-f8v2-person-mask2former-swinb-v4") model = Mask2FormerForUniversalSegmentation.from_pretrained("james-emi/darpa_ir-f8v2-person-mask2former-swinb-v4") - Notebooks
- Google Colab
- Kaggle
| { | |
| "activation_function": "relu", | |
| "architectures": [ | |
| "Mask2FormerForUniversalSegmentation" | |
| ], | |
| "backbone": null, | |
| "backbone_config": { | |
| "attention_probs_dropout_prob": 0.0, | |
| "depths": [ | |
| 2, | |
| 2, | |
| 18, | |
| 2 | |
| ], | |
| "drop_path_rate": 0.3, | |
| "embed_dim": 128, | |
| "encoder_stride": 32, | |
| "hidden_act": "gelu", | |
| "hidden_dropout_prob": 0.0, | |
| "hidden_size": 1024, | |
| "image_size": 224, | |
| "initializer_range": 0.02, | |
| "layer_norm_eps": 1e-05, | |
| "mlp_ratio": 4.0, | |
| "model_type": "swin", | |
| "num_channels": 3, | |
| "num_heads": [ | |
| 4, | |
| 8, | |
| 16, | |
| 32 | |
| ], | |
| "num_layers": 4, | |
| "out_features": [ | |
| "stage1", | |
| "stage2", | |
| "stage3", | |
| "stage4" | |
| ], | |
| "out_indices": [ | |
| 1, | |
| 2, | |
| 3, | |
| 4 | |
| ], | |
| "patch_size": 4, | |
| "path_norm": true, | |
| "qkv_bias": true, | |
| "stage_names": [ | |
| "stem", | |
| "stage1", | |
| "stage2", | |
| "stage3", | |
| "stage4" | |
| ], | |
| "torch_dtype": "float32", | |
| "use_absolute_embeddings": false, | |
| "window_size": 12 | |
| }, | |
| "backbone_kwargs": null, | |
| "class_weight": 2.0, | |
| "common_stride": 4, | |
| "decoder_layers": 10, | |
| "dice_weight": 5.0, | |
| "dim_feedforward": 2048, | |
| "dropout": 0.0, | |
| "encoder_feedforward_dim": 1024, | |
| "encoder_layers": 6, | |
| "enforce_input_proj": false, | |
| "enforce_input_projection": false, | |
| "feature_size": 256, | |
| "feature_strides": [ | |
| 4, | |
| 8, | |
| 16, | |
| 32 | |
| ], | |
| "hidden_dim": 256, | |
| "id2label": { | |
| "0": "person" | |
| }, | |
| "ignore_value": 255, | |
| "importance_sample_ratio": 0.75, | |
| "init_std": 0.02, | |
| "init_xavier_std": 1.0, | |
| "label2id": { | |
| "person": 0 | |
| }, | |
| "mask_feature_size": 256, | |
| "mask_weight": 5.0, | |
| "model_type": "mask2former", | |
| "no_object_weight": 0.1, | |
| "num_attention_heads": 8, | |
| "num_hidden_layers": 10, | |
| "num_queries": 100, | |
| "output_auxiliary_logits": null, | |
| "oversample_ratio": 3.0, | |
| "pre_norm": false, | |
| "torch_dtype": "float32", | |
| "train_num_points": 12544, | |
| "transformers_version": "4.55.1", | |
| "use_auxiliary_loss": true, | |
| "use_pretrained_backbone": false, | |
| "use_timm_backbone": false | |
| } | |