Adding all required files for NEW QUESTION : IoU Score Calculation

questions/_template/description.md

@@ -1,3 +1,3 @@
-## Problem
+Write a Python function to implement the Intersection over Union Score. The function should calculate the score of area of overlap of 2 bounding boxes divided by the area of union of both box. 
 
-Write a concise problem description here.
+The output : iou score of datatype float

questions/_template/example.json

@@ -1,5 +1,14 @@
 {
-  "input": "...",
-  "output": "...",
-  "reasoning": "Explain why the output follows from the input."
+  "input": {
+    "bbox_A": [100, 100, 200, 200],
+    "bbox_B": [150, 150, 250, 250]
+  },
+  "output": {
+    "iou_score": 0.146
+  },
+  "reasoning": "The output shows the Intersection over Union (IoU) score between bbox_A and bbox_B. Here's how it's calculated: bbox_A is [100,100,200,200] representing [x1,y1,x2,y2] bbox_B is [150,150,250,250] representing [x1,y1,x2,y2]
+                The intersection area is 50x50=2500 (overlapping region).The union area is the sum of both areas minus the intersection:   
+                - Area_A = 100x100 = 10000  - Area_B = 100x100 = 10000
+                - Union = 10000 + 10000 - 2500 = 17500  IoU = intersection/union = 2500/17500 ≈ 0.14286
+              "
 }

questions/_template/learn.md

@@ -1,47 +1,64 @@
 ## Solution Explanation
 
-Add intuition, math, and step-by-step reasoning here.
-
-### Writing Mathematical Expressions with LaTeX
-
-This editor supports LaTeX for rendering mathematical equations and expressions. Here's how you can use it:
-
-1. **Inline Math**: 
-   - Wrap your expression with single `$` symbols.
-   - Example: `$E = mc^2$` → Renders as: ( $E = mc^2$ )
-
-2. **Block Math**:
-   - Wrap your expression with double `$$` symbols.
-   - Example: 
-     ```
-     $$
-     \int_a^b f(x) \, dx
-     $$
-     ```
-     Renders as:
-     $$
-     \int_a^b f(x) \, dx
-     $$
-
-3. **Math Functions**:
-   - Use standard LaTeX functions like `\frac`, `\sqrt`, `\sum`, etc.
-   - Examples:
-     - `$\frac{a}{b}$` → ( $\frac{a}{b}$ )
-     - `$\sqrt{x}$` → ( $\sqrt{x}$ )
-
-4. **Greek Letters and Symbols**:
-   - Use commands like `\alpha`, `\beta`, etc., for Greek letters.
-   - Example: `$\alpha + \beta = \gamma$` → ( $\alpha + \beta = \gamma$ )
-
-5. **Subscripts and Superscripts**:
-   - Use `_{}` for subscripts and `^{}` for superscripts.
-   - Examples:
-     - `$x_i$` → ( $x_i$ )
-     - `$x^2$` → ( $x^2$ )
-
-6. **Combined Examples**:
-   - `$\sum_{i=1}^n i^2 = \frac{n(n+1)(2n+1)}{6}$`
-     Renders as:
-     $\sum_{i=1}^n i^2 = \frac{n(n+1)(2n+1)}{6}$
-
-Feel free to write your own mathematical expressions, and they will be rendered beautifully in the preview!
+Intuition
+----
+
+The Intersection over Union (IoU) score is a metric used to evaluate how well a predicted region (for example, a bounding box or segmentation mask) matches the ground truth region.
+
+The intuition is simple:
+
+It measures how much the predicted area overlaps with the true area,
+And compares that overlap to the total area covered by both regions combined.
+If the prediction perfectly matches the ground truth, IoU = 1 (perfect overlap).
+If the prediction does not overlap at all, IoU = 0.
+
+Thus, IoU quantifies spatial similarity between the prediction and the ground truth.
+
+Mathematical definition:
+IoU = Area of Overlap/Area of Union
+
+
+## Step-by-step reasoning: 
+
+1. Identify regions: Determine the predicted area B and ground truth area A.
+
+    `INPUTS bboxA and bboxB`
+
+    `bbox = [x1, y1, x2, y2]` 
+
+    (x1,y1 -> top left coordinates || x2,y2 -> bottom right coordinates )
+
+
+2. Compute intersection: Calculate the overlapping area  `|A ∩ B∣`
+
+3. Compute union: Calculate the combined area ∣ 𝐴 ∪ 𝐵 ∣ = ∣ 𝐴 ∣ + ∣ 𝐵 ∣ − ∣ 𝐴 ∩ 𝐵 ∣
+` ∣ A∪B ∣ = ∣A∣ + ∣B∣ − ∣ A∩B∣ `
+
+4. Compute IoU: Divide the intersection area by the union area.
+
+
+Example:
+If  
+∣A∣=100, ∣B∣=80, and ∣A∩B∣=60,
+then:
+
+`∣A∪B∣ = 100 + 80 − 60 = 120`
+
+`IoU = 60/120 = 0.5`
+
+→ The prediction overlaps 50% with the ground truth.
+
+
+## THE MATH  
+
+The IoU score is calculated as $IoU = \frac{|A \cap B|}{|A \cup B|}$,
+where $A$ is the ground truth and $B$ is the predicted region.
+
+where:  
+A = Ground truth region   
+B = Predicted region   
+∣A∩B∣ = Intersection area (common area between A and B)  
+∣A∪B∣ = Union area (total area covered by A and B)
+
+
+--------

questions/_template/meta.json

@@ -1,12 +1,17 @@
 {
   "id": "XXX",
-  "title": "TITLE GOES HERE",
+  "title": "IOU-score calculation",
   "difficulty": "medium",
-  "category": "Machine Learning",
+  "category": "Computer Vision",
   "video": "",
   "likes": "0",
   "dislikes": "0",
-  "contributor": [],
+  "contributor": [
+    {
+      "profile_link": "https://github.com/dsaha21",
+      "name": "Dripto Saha"
+    }
+  ],
   "tinygrad_difficulty": "",
   "pytorch_difficulty": ""
 }

questions/_template/solution.py

@@ -1,3 +1,17 @@
-def your_function(...):
-    # reference implementation
-    ...
+import numpy as np
+def iou_score(bboxA, bboxB):
+    xA = np.maximum(bboxA[0], bboxB[0])
+    yA = np.maximum(bboxA[1], bboxB[1])
+    xB = np.minimum(bboxA[2], bboxB[2])
+    yB = np.minimum(bboxA[3], bboxB[3])
+
+    interWidth = np.maximum(0, xB - xA + 1)
+    interHeight = np.maximum(0, yB - yA + 1)
+    interArea = interWidth * interHeight
+
+    boxAArea = (bboxA[2] - bboxA[0] + 1) * (bboxA[3] - bboxA[1] + 1)
+    boxBArea = (bboxB[2] - bboxB[0] + 1) * (bboxB[3] - bboxB[1] + 1)
+
+    #unionArea = boxAArea + boxBArea - interArea
+    iou = interArea / float(boxAArea + boxBArea - interArea)
+    return round(iou, 3)

questions/_template/starter_code.py

@@ -1,4 +1,12 @@
 # Implement your function below.
 
-def your_function(...):
-    pass
+def iou_score(bboxA, bboxB):
+    #xA = np.maximum(bboxA[0], bboxB[0])
+
+    #interWidth = np.maximum(0, xB - xA + 1)
+    #interArea = interWidth * interHeight
+
+    #find bbox areas of A and B
+
+    iou = interArea / float(boxAArea + boxBArea - interArea)
+    return round(iou, 3)

questions/_template/tests.json

@@ -1,6 +1,24 @@
 [
   {
-    "test": "print(your_function(...))",
-    "expected_output": "..."
+    "test": "print(iou_score([0, 0, 100, 100],[0, 0, 100, 100]))", 
+    "expected_output": "1.0"
+  },
+  {
+    "test": "print(iou_score([0, 0, 50, 50],[60, 60, 100, 100]))", 
+    "expected_output": "0.0"
+  },
+  {
+    "test": "print(iou_score([100, 100, 200, 200],[150, 150, 250, 250]))", 
+    "expected_output": "0.146"
+  },
+  {
+    "test": "print(iou_score([10, 20, 100, 100],[90, 90, 200, 201]))", 
+    "expected_output": "0.006"
+  },
+  {
+    "test": "print(iou_score([100, 100.2, 200.7, 201.5],[150.5, 150, 250.8, 250.9]))", 
+    "expected_output": "0.149"
   }
+
+
 ]