Upload 10 files

.gitattributes

@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+images/baklava.png filter=lfs diff=lfs merge=lfs -text
+images/beignets.png filter=lfs diff=lfs merge=lfs -text
+images/cat.png filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,48 @@
+# Necessary imports
+import gradio as gr
+from src.siglip.classifier import ZeroShotImageClassification
+
+
+# Examples to display in the interface
+examples = [
+    [
+        "images/baklava.png",
+        "dessert on a plate, a serving of baklava, a plate and spoon",
+    ],
+    [
+        "images/beignets.png",
+        "a dog, a cat, a donut, a beignet",
+    ],
+    [
+        "images/cat.png",
+        "two sleeping cats, two cats playing, three cats laying down",
+    ],
+]
+
+# Title and description and article for the interface
+title = "Zero Shot Image Classification"
+description = "Classify image using zero-shot classification with SigLIP 2 zeroshot model! Provide an image input and a list of candidate labels separated by commas. Read more at the links below."
+article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2502.14786' target='_blank'>SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features</a> | <a href='https://huggingface.co/google/siglip2-so400m-patch16-naflex' target='_blank'>Model Page</a></p>"
+
+
+# Launch the interface
+demo = gr.Interface(
+    fn=ZeroShotImageClassification,
+    inputs=[
+        gr.Image(type="pil", label="Input", placeholder="Enter image to classify"),
+        gr.Textbox(
+            label="Candidate Labels",
+            placeholder="Enter candidate labels separated by commas",
+        ),
+    ],
+    outputs=gr.Label(label="Classification"),
+    title=title,
+    description=description,
+    article=article,
+    examples=examples,
+    cache_examples=True,
+    cache_mode="lazy",
+    theme="Soft",
+    flagging_mode="never",
+)
+demo.launch(debug=False)

requirements.txt

@@ -0,0 +1,7 @@
+gradio
+torch
+git+https://github.com/huggingface/transformers@main
+sentencepiece
+pillow
+protobuf
+accelerate

src/exception.py

@@ -0,0 +1,50 @@
+"""
+This module defines a custom exception handling class and a function to get error message with details of the error.
+"""
+
+# Standard Library
+import sys
+
+# Local imports
+from src.logger import logging
+
+
+# Function Definition to get error message with details of the error (file name and line number) when an error occurs in the program
+def get_error_message(error, error_detail: sys):
+    """
+    Get error message with details of the error.
+
+    Args:
+        - error (Exception): The error that occurred.
+        - error_detail (sys): The details of the error.
+
+    Returns:
+        str: A string containing the error message along with the file name and line number where the error occurred.
+    """
+    _, _, exc_tb = error_detail.exc_info()
+
+    # Get error details
+    file_name = exc_tb.tb_frame.f_code.co_filename
+    return "Error occured in python script name [{0}] line number [{1}] error message[{2}]".format(
+        file_name, exc_tb.tb_lineno, str(error)
+    )
+
+
+# Custom Exception Handling Class Definition
+class CustomExceptionHandling(Exception):
+    """
+    Custom Exception Handling:
+        This class defines a custom exception that can be raised when an error occurs in the program.
+        It takes an error message and an error detail as input and returns a formatted error message when the exception is raised.
+    """
+
+    # Constructor
+    def __init__(self, error_message, error_detail: sys):
+        """Initialize the exception"""
+        super().__init__(error_message)
+
+        self.error_message = get_error_message(error_message, error_detail=error_detail)
+
+    def __str__(self):
+        """String representation of the exception"""
+        return self.error_message

src/logger.py

@@ -0,0 +1,21 @@
+# Importing the required modules
+import os
+import logging
+from datetime import datetime
+
+# Creating a log file with the current date and time as the name of the file
+LOG_FILE = f"{datetime.now().strftime('%m_%d_%Y_%H_%M_%S')}.log"
+
+# Creating a logs folder if it does not exist
+logs_path = os.path.join(os.getcwd(), "logs", LOG_FILE)
+os.makedirs(logs_path, exist_ok=True)
+
+# Setting the log file path and the log level
+LOG_FILE_PATH = os.path.join(logs_path, LOG_FILE)
+
+# Configuring the logger
+logging.basicConfig(
+    filename=LOG_FILE_PATH,
+    format="[ %(asctime)s ] %(lineno)d %(name)s - %(levelname)s - %(message)s",
+    level=logging.INFO,
+)

src/siglip/classifier.py

@@ -0,0 +1,63 @@
+# Necessary imports
+import sys
+from typing import Dict
+import torch
+from transformers import AutoModel, AutoProcessor
+import gradio as gr
+
+# Local imports
+from src.logger import logging
+from src.exception import CustomExceptionHandling
+
+
+# Load the zero-shot image classification model
+model_id = "google/siglip2-so400m-patch16-naflex"
+model = AutoModel.from_pretrained(model_id, device_map="cpu").eval()
+processor = AutoProcessor.from_pretrained(model_id)
+
+
+def ZeroShotImageClassification(
+    image_input: str, candidate_labels: str
+) -> Dict[str, float]:
+    """
+    Performs zero-shot classification on the given image input and candidate labels.
+
+    Args:
+        - image_input: The input image to classify.
+        - candidate_labels: A comma-separated string of candidate labels.
+
+    Returns:
+        Dictionary containing label-score pairs.
+    """
+    try:
+        # Check if the input and candidate labels are valid
+        if not image_input or not candidate_labels:
+            gr.Warning("Please provide valid input and candidate labels")
+
+        # Split and clean the candidate labels
+        labels = [label.strip() for label in candidate_labels.split(",")]
+
+        # Log the classification attempt
+        logging.info(f"Attempting classification with {len(labels)} labels")
+
+        # Perform zero-shot image classification
+        inputs = processor(
+            text=labels,
+            images=image_input,
+            return_tensors="pt",
+            padding="max_length",
+            max_length=64,
+        ).to("cpu")
+        with torch.no_grad():
+            outputs = model(**inputs)
+            logits_per_image = outputs.logits_per_image
+            probs = torch.sigmoid(logits_per_image)
+
+        # Return the classification results
+        logging.info("Classification completed successfully")
+        return {labels[i]: probs[0][i] for i in range(len(labels))}
+
+    # Handle exceptions that may occur during the process
+    except Exception as e:
+        # Custom exception handling
+        raise CustomExceptionHandling(e, sys) from e