Instructions to use zeromodels/rtdetr-r50vd-coco-o365 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ZeroModels
How to use zeromodels/rtdetr-r50vd-coco-o365 with ZeroModels:
# pip install -U zeromodels # ZeroModels is pure Keras 3, so pick a backend: "jax", "torch" or "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" from zeromodels import AutoZModel # AutoZModel reads the repo's model_type and loads the matching class. # For a task head use the matching loader, e.g. AutoZMImageClassify / AutoZMDetect / # AutoZMSemanticSegment / AutoZMTextGenerate (see zeromodels.auto). model = AutoZModel.from_weights("zeromodels/rtdetr-r50vd-coco-o365") - Keras
How to use zeromodels/rtdetr-r50vd-coco-o365 with Keras:
# Available backend options are: "jax", "torch", "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" import keras model = keras.saving.load_model("hf://zeromodels/rtdetr-r50vd-coco-o365") - Notebooks
- Google Colab
- Kaggle
Migrate to zeromodels (rename kf_*.json -> zm_*.json, fix refs in config + README, ensure tag + badge)
9d58440 verified | { | |
| "library_name": "zeromodels", | |
| "zeromodels_version": "1.2.1", | |
| "model_module": "zeromodels.models.rt_detr", | |
| "model_class": "RTDETRDetect", | |
| "variant": "rtdetr-r50vd-coco-o365", | |
| "weights": "model.weights.h5", | |
| "schema_version": 2, | |
| "weight_dtype": "float32", | |
| "model_type": "rt_detr", | |
| "vision_config": { | |
| "backbone_hidden_sizes": [ | |
| 256, | |
| 512, | |
| 1024, | |
| 2048 | |
| ], | |
| "backbone_block_repeats": [ | |
| 3, | |
| 4, | |
| 6, | |
| 3 | |
| ], | |
| "backbone_embedding_size": 64, | |
| "backbone_layer_type": "bottleneck", | |
| "encoder_in_channels": [ | |
| 512, | |
| 1024, | |
| 2048 | |
| ], | |
| "encoder_hidden_dim": 256, | |
| "encoder_num_layers": 1, | |
| "encoder_ffn_dim": 1024, | |
| "encoder_num_heads": 8, | |
| "encode_proj_layers": [ | |
| 2 | |
| ], | |
| "encoder_activation_function": "gelu", | |
| "activation_function": "silu", | |
| "hidden_expansion": 1.0, | |
| "hidden_dim": 256, | |
| "decoder_num_layers": 6, | |
| "decoder_ffn_dim": 1024, | |
| "decoder_num_heads": 8, | |
| "decoder_n_points": 4, | |
| "decoder_activation_function": "relu", | |
| "num_feature_levels": 3, | |
| "feat_strides": [ | |
| 8, | |
| 16, | |
| 32 | |
| ], | |
| "num_queries": 300, | |
| "num_classes": 80, | |
| "image_size": 640 | |
| } | |
| } |