Instructions to use zeromodels/sam2_hiera_base_plus with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ZeroModels
How to use zeromodels/sam2_hiera_base_plus 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/sam2_hiera_base_plus") - Keras
How to use zeromodels/sam2_hiera_base_plus with Keras:
# !pip install -U keras tensorflow huggingface_hub # Keras needs TensorFlow installed to read "hf://" paths, so the tensorflow backend is selected here; # "jax" and "torch" also work for computation once TensorFlow is installed. import os os.environ["KERAS_BACKEND"] = "tensorflow" import keras model = keras.saving.load_model("hf://zeromodels/sam2_hiera_base_plus") - sam2
How to use zeromodels/sam2_hiera_base_plus with sam2:
# Use SAM2 with images import torch from sam2.sam2_image_predictor import SAM2ImagePredictor predictor = SAM2ImagePredictor.from_pretrained("zeromodels/sam2_hiera_base_plus") with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): predictor.set_image(<your_image>) masks, _, _ = predictor.predict(<input_prompts>)# Use SAM2 with videos import torch from sam2.sam2_video_predictor import SAM2VideoPredictor predictor = SAM2VideoPredictor.from_pretrained("zeromodels/sam2_hiera_base_plus") with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): state = predictor.init_state(<your_video>) # add new prompts and instantly get the output on the same frame frame_idx, object_ids, masks = predictor.add_new_points(state, <your_prompts>) # propagate the prompts to get masklets throughout the video for frame_idx, object_ids, masks in predictor.propagate_in_video(state): ... - Notebooks
- Google Colab
- Kaggle
Upgrade kf_config.json to v2 (nested config + weight_dtype)
Browse files- kf_config.json +1 -1
kf_config.json
CHANGED
|
@@ -6,7 +6,7 @@
|
|
| 6 |
"variant": "sam2_hiera_base_plus",
|
| 7 |
"weights": "model.weights.h5",
|
| 8 |
"schema_version": 2,
|
| 9 |
-
"
|
| 10 |
"model_type": "sam2",
|
| 11 |
"vision_config": {
|
| 12 |
"hidden_dim": 112,
|
|
|
|
| 6 |
"variant": "sam2_hiera_base_plus",
|
| 7 |
"weights": "model.weights.h5",
|
| 8 |
"schema_version": 2,
|
| 9 |
+
"weight_dtype": "float32",
|
| 10 |
"model_type": "sam2",
|
| 11 |
"vision_config": {
|
| 12 |
"hidden_dim": 112,
|