Tabular Regression
Transformers
Safetensors
yield-text-image
feature-extraction
crop-yield-estimation
digital-agriculture
multimodal
multi-temporal
regression
custom_code
Instructions to use ICICLE-AI/multimodal-yield-estimation with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ICICLE-AI/multimodal-yield-estimation with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("ICICLE-AI/multimodal-yield-estimation", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download configuration_multimodal.py from ICICLE-AI/multimodal-yield-estimation: direct link, hf CLI and curl.
- Browser
- Download file 1.62 kB
-
https://huggingface.co/ICICLE-AI/multimodal-yield-estimation/resolve/main/configuration_multimodal.py
- Command line
-
hf download hf://ICICLE-AI/multimodal-yield-estimation/configuration_multimodal.py
-
curl -L -o configuration_multimodal.py https://huggingface.co/ICICLE-AI/multimodal-yield-estimation/resolve/main/configuration_multimodal.py
1.62 kB
| from transformers import PretrainedConfig | |
| class MultimodalYieldConfig(PretrainedConfig): | |
| """Config for the text + image (concat) multi-cutoff yield model. | |
| Normalization statistics are stored here so the model accepts raw | |
| Qwen3 text embeddings and raw DINOv3 image features at inference time. | |
| """ | |
| model_type = "yield-text-image" | |
| def __init__( | |
| self, | |
| text_model="Qwen/Qwen3-Embedding-0.6B", | |
| image_model="facebook/dinov3-vitb16-pretrain-lvd1689m", | |
| text_dim=1024, | |
| image_dim=768, | |
| K=22, | |
| text_mean=None, | |
| text_std=None, | |
| image_mean=None, | |
| image_std=None, | |
| y_mean=None, | |
| y_std=None, | |
| train_cutoffs=None, | |
| eval_cutoffs=None, | |
| d_model=64, | |
| nhead=4, | |
| num_layers=2, | |
| dim_ff=128, | |
| dropout=0.3, | |
| input_dropout=0.2, | |
| pool="last", | |
| **kwargs, | |
| ): | |
| super().__init__(**kwargs) | |
| self.text_model = text_model | |
| self.image_model = image_model | |
| self.text_dim = text_dim | |
| self.image_dim = image_dim | |
| self.K = K | |
| self.text_mean = text_mean | |
| self.text_std = text_std | |
| self.image_mean = image_mean | |
| self.image_std = image_std | |
| self.y_mean = y_mean | |
| self.y_std = y_std | |
| self.train_cutoffs = train_cutoffs | |
| self.eval_cutoffs = eval_cutoffs | |
| self.d_model = d_model | |
| self.nhead = nhead | |
| self.num_layers = num_layers | |
| self.dim_ff = dim_ff | |
| self.dropout = dropout | |
| self.input_dropout = input_dropout | |
| self.pool = pool | |