init

README.md

@@ -1,13 +1,26 @@
----
-title: Bioclip Canopy
-emoji: 🦀
-colorFrom: green
-colorTo: indigo
-sdk: gradio
-sdk_version: 4.37.2
-app_file: app.py
-pinned: false
-license: apache-2.0
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# BioClip Image Classification
+
+This Hugging Face Space demonstrates image classification using the BioClip model. Upload an image to get a prediction of its class, along with the top 3 most similar classes and file paths.
+
+## How to Use
+
+1. Open the Gradio interface in this Space.
+2. Upload an image using the provided input area.
+3. The model will process the image and return:
+   - The predicted class
+   - The top 3 most similar classes
+   - The top 3 most similar file paths from the dataset
+
+## About the Model
+
+This Space uses the BioClip model, which is designed for biological image classification. The model is loaded from the Hugging Face model hub (imageomics/bioclip).
+
+## Technical Details
+
+- The Space uses Gradio for the user interface.
+- It employs FAISS indexes for efficient similarity search.
+- The classification is performed using a k-nearest neighbors approach with majority voting.
+
+## Note
+
+The dataset and FAISS indexes are expected to be present in the `./data/embeddings_bioclip_False` directory. Make sure to include these files when setting up the Space.

app.py

@@ -0,0 +1,78 @@
+import gradio as gr
+import numpy as np
+import torch
+from PIL import Image
+import open_clip
+from datasets import Dataset
+import os
+
+# Set environment variable to work around OpenMP runtime issue
+os.environ['KMP_DUPLICATE_LIB_OK'] = 'TRUE'
+
+# Load the model and processor
+model, processor = open_clip.create_model_from_pretrained('hf-hub:imageomics/bioclip')
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model.to(device)
+
+# Load the dataset
+embedding_path = "./data/embeddings_bioclip_False"
+ds = Dataset.load_from_disk(embedding_path)
+
+# Load FAISS indexes
+cosine_faiss_path = os.path.join(embedding_path, "embeddings_cosine.faiss")
+l2_faiss_path = os.path.join(embedding_path, "embeddings_l2.faiss")
+ds.load_faiss_index("embeddings_cosine", cosine_faiss_path)
+ds.load_faiss_index("embeddings_l2", l2_faiss_path)
+
+def majority_vote(classes, scores=None):
+    if scores is None:
+        scores = np.ones_like(classes)
+    unique_classes, class_counts = np.unique(classes, return_counts=True)
+    class_weights = {cls: 0 for cls in unique_classes}
+
+    for cls, weight in zip(classes, scores):
+        class_weights[cls] += weight
+
+    majority_class = max(class_weights, key=class_weights.get)
+    return majority_class
+
+def classify_example(example, index="embeddings_l2", k=10, vote_scores=True):
+    features = np.array(example["embeddings"], dtype=np.float32)
+    scores, nearest = ds.get_nearest_examples(index, features, k)
+
+    class_labels = [ds.features["label"].names[c] for c in nearest["label"]]
+
+    if vote_scores:
+        prediction = majority_vote(class_labels, scores)
+    else:
+        prediction = majority_vote(class_labels)
+    
+    return prediction, class_labels, nearest["file"]
+
+def embed_image(image: Image.Image):
+    processed_images = processor(image).unsqueeze(0)
+
+    with torch.no_grad():
+        embeddings = model.encode_image(processed_images.to(device))
+
+    return {"embeddings": embeddings.cpu()}
+
+def predict(image):
+    embedding = embed_image(image)
+    prediction, class_labels, file_paths = classify_example(embedding)
+    
+    return prediction, ", ".join(class_labels[:3]), ", ".join(file_paths[:3])
+
+iface = gr.Interface(
+    fn=predict,
+    inputs=gr.Image(type="pil"),
+    outputs=[
+        gr.Textbox(label="Prediction"),
+        gr.Textbox(label="Top 3 Classes"),
+        gr.Textbox(label="Top 3 File Paths")
+    ],
+    title="BioClip Image Classification",
+    description="Upload an image to get a prediction using the BioClip model."
+)
+
+iface.launch()

requirements.txt

@@ -0,0 +1,7 @@
+gradio
+numpy
+torch
+Pillow
+open_clip_torch
+datasets
+faiss-cpu