Create README.md

unimodal-sss-bnn/README.md

@@ -0,0 +1,98 @@
+---
+
+## Unimodal SSS Bayesian Neural Network Classifier - Model Card
+
+Here is the Markdown content for `sams-tom/multimodal-auv-bathy-bnn-classifier/unimodal-sss-bnn/README.md`:
+
+```markdown
+---
+license: mit
+language:
+- en
+metrics:
+- accuracy
+base_model:
+- microsoft/resnet-50
+pipeline_tag: image-classification
+tags:
+- AUV
+- biology
+- marine
+- sonar
+- SSS
+- unimodal
+- ecology
+- robotics
+---
+
+# Model Card: Unimodal SSS Bayesian Neural Network Classifier
+
+This model is a Bayesian Neural Network (BNN) classifier, specifically a `ResNet50Custom` implementation, designed for the classification of AUV side-scan sonar (SSS) data. It utilizes a ResNet50 backbone, pre-trained on ImageNet, adapted for 1-channel SSS inputs. As a BNN, it provides uncertainty estimates alongside its predictions, which can be valuable for decision-making in uncertain underwater environments.
+
+---
+
+## Model Details
+
+### Model Description
+
+This model is a Bayesian Neural Network (BNN) classifier, specifically a `ResNet50Custom` implementation, specialized for the classification of **AUV side-scan sonar (SSS) data**. It leverages a **ResNet50** backbone (pre-trained on ImageNet-1K) as its primary feature extractor. The model's input layer (`conv1`) has been adapted to handle 1 channel, making it suitable for grayscale SSS waterfall plots.
+
+As a **Bayesian Neural Network**, it provides valuable uncertainty estimates alongside its class predictions, making it suitable for applications where quantifying confidence in predictions is important.
+
+* **Developed by:** Sams-Tom
+* **Shared by:** Sams-Tom
+* **Model type:** Bayesian Neural Network (BNN), Side-Scan Sonar Classifier (Computer Vision)
+* **Language(s) (NLP):** N/A (Not an NLP model)
+* **License:** MIT
+* **Finetuned from model [optional]:** ResNet50 (pre-trained on ImageNet-1K)
+
+### Model Sources
+
+* **Repository:** `https://huggingface.co/sams-tom/multimodal-auv-bathy-bnn-classifier`
+* **Paper:** [In development]
+* **Demo [optional]:** [More Information Needed: If there's a live demo or notebook, please link it here.]
+
+
+---
+
+## How to Get Started with the Model
+
+To use this model, ensure you have `torch`, `huggingface_hub`, and `bayesian-torch` installed (`pip install torch huggingface_hub bayesian-torch`).
+
+```python
+from transformers import AutoModel
+import torch
+import json
+import os
+from huggingface_hub import hf_hub_download
+
+# Assuming your custom model definitions are accessible (e.g., in model_definitions.py in repo root)
+
+repo_id = "sams-tom/multimodal-auv-bathy-bnn-classifier"
+model_subfolder = "unimodal-sss-bnn"
+
+# Load the BNN prior parameters that were used during training/conversion.
+# These are essential for understanding the model's Bayesian properties.
+bnn_params_path = hf_hub_download(repo_id=repo_id, filename=os.path.join(model_subfolder, "bnn_params.json"))
+with open(bnn_params_path, "r") as f:
+    const_bnn_prior_parameters = json.load(f)
+print(f"Loaded BNN Prior Parameters: {const_bnn_prior_parameters}")
+
+# Load the model directly using AutoModel, specifying the subfolder.
+# trust_remote_code=True is crucial because your model uses custom classes (ResNet50Custom etc.)
+# which are defined in the 'model_definitions.py' file uploaded to the root of the repository.
+model = AutoModel.from_pretrained(repo_id, subfolder=model_subfolder, trust_remote_code=True)
+model.eval() # Set to evaluation mode
+
+# Example inference with dummy SSS data
+# Adjust dimensions (batch size, channels, height, width) based on your actual data
+dummy_sss_input = torch.randn(1, 1, 224, 224) # Example: (Batch, Grayscale Channel, Height, Width)
+
+with torch.no_grad():
+    outputs = model(dummy_sss_input)
+    probabilities = torch.softmax(outputs, dim=-1) # Get class probabilities
+    predicted_class = torch.argmax(probabilities, dim=-1) # Get predicted class index
+
+print(f"Output logits: {outputs}")
+print(f"Predicted class probabilities: {probabilities}")
+print(f"Predicted class index: {predicted_class.item()}")