Explainable AI (XAI) and Grad-CAM

19. Explainable AI (XAI) and Grad-CAM#

19.1. 1. Introduction to Explainable AI (XAI)#

Artificial Intelligence (AI) has become ubiquitous in many industries, from healthcare to finance, where decisions made by AI systems can have significant impacts. However, many AI models, especially deep learning models, are often considered “black boxes” because they do not provide insights into how they arrive at their predictions. This lack of transparency leads to a critical question:

How do we trust AI systems if we cannot understand their decision-making process?

Explainable AI (XAI) seeks to address this issue by providing methods to interpret and explain the decisions of AI models. With XAI, we can:

Increase trust and accountability of AI models
Detect biases or potential errors in predictions
Comply with regulatory requirements, especially in sensitive domains like healthcare and finance

We will explore one such method: Class Activation Mapping (CAM), particularly Gradient-weighted Class Activation Mapping (Grad-CAM), which is used for explaining the predictions of Convolutional Neural Networks (CNNs).

19.2. 2. What is Class Activation Mapping (CAM)?#

Class Activation Mapping (CAM) is a technique used to highlight regions in an input image that are important for the CNN’s prediction of a specific class. This method essentially provides visual explanations by generating a heatmap over the input image, showing the areas that contributed most to the final prediction.

Originally, CAM required some modifications to the CNN architecture, such as the addition of global average pooling layers. However, Grad-CAM extends CAM to work with any CNN architecture by leveraging the gradients of the predicted class with respect to the feature maps in the convolutional layers.

19.2.1. Grad-CAM Overview#

Grad-CAM uses the gradients flowing into the last convolutional layer to assign importance values to each neuron for a particular decision. In essence, it creates a coarse localization map highlighting important regions in the image for a specific class prediction. The original version of Grad-CAM can be found in arxiv:1610.02391.

19.2.2. Steps to compute Grad-CAM:#

Forward Pass: Perform a forward pass through the network to obtain the class scores (logits) for the input image.
Compute Gradients: Compute the gradients (\(y^{k}\)) of the predicted class score with respect to the feature maps of the last convolutional layer (\(k\)) i.e. \(\frac{\delta y^{k}}{\delta A_{ij}^{k}}\)
Weighted Feature Map: The gradients are globally averaged to obtain the importance weights. Global average pool over the width dimension (\(i\)) and the height dimension (\(j\)) to obtain the weights \(w_{k}^{c}\)

\[ w_{k}^{c} = \frac{1}{Z} \sum_{i}\sum_{j} \frac{\delta y^{c}}{ \delta A_{ij}^{k} } \]

**Calculate Final Grad-CAM localisation map **

Perform a weighted combination of the feature map activations \(A^{k}\) where the weights are the \(w_{k}^{c}\) we just calculate the positive contributions by applying a ReLU function.

\[ L_{\text{Grad-CAM}}^{c} = \text{ReLU}(\sum_{k} w_{k}^{c} A^{k}) \]

Generate Heatmap: These weights are combined with the feature maps to produce a heatmap, highlighting the areas of the input image most relevant to the prediction.
Overlay on Input: Finally, the heatmap is superimposed on the input image for visualization, For this step, the heatmap has to be rescaled to the size of the input.

19.3. 3. Why Grad-CAM for CNN Explainability?#

CNNs, being hierarchical in nature, learn abstract features as we go deeper into the network. Grad-CAM provides insight into what part of the image the model focuses on while making a prediction, allowing us to visually understand the model’s reasoning. This can be especially useful in applications like:

Medical Imaging: Understanding which regions of an image are associated with a diagnosis. Grad-CAM for DL models classify multiple sclerosis types using brain MRI
Autonomous Driving: Checking whether the network is paying attention to relevant objects, like pedestrians or traffic signs. vechicle steering angle visualization
Object Classification: Ensuring that the model focuses on the correct part of the image for classifications

!pip install torch torchvision matplotlib

Requirement already satisfied: torch in /usr/local/lib/python3.10/dist-packages (2.4.1+cu121)
Requirement already satisfied: torchvision in /usr/local/lib/python3.10/dist-packages (0.19.1+cu121)
Requirement already satisfied: matplotlib in /usr/local/lib/python3.10/dist-packages (3.7.1)
Requirement already satisfied: filelock in /usr/local/lib/python3.10/dist-packages (from torch) (3.16.1)
Requirement already satisfied: typing-extensions>=4.8.0 in /usr/local/lib/python3.10/dist-packages (from torch) (4.12.2)
Requirement already satisfied: sympy in /usr/local/lib/python3.10/dist-packages (from torch) (1.13.3)
Requirement already satisfied: networkx in /usr/local/lib/python3.10/dist-packages (from torch) (3.4)
Requirement already satisfied: jinja2 in /usr/local/lib/python3.10/dist-packages (from torch) (3.1.4)
Requirement already satisfied: fsspec in /usr/local/lib/python3.10/dist-packages (from torch) (2024.6.1)
Requirement already satisfied: numpy in /usr/local/lib/python3.10/dist-packages (from torchvision) (1.26.4)
Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /usr/local/lib/python3.10/dist-packages (from torchvision) (10.4.0)
Requirement already satisfied: contourpy>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (1.3.0)
Requirement already satisfied: cycler>=0.10 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (0.12.1)
Requirement already satisfied: fonttools>=4.22.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (4.54.1)
Requirement already satisfied: kiwisolver>=1.0.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (1.4.7)
Requirement already satisfied: packaging>=20.0 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (24.1)
Requirement already satisfied: pyparsing>=2.3.1 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (3.1.4)
Requirement already satisfied: python-dateutil>=2.7 in /usr/local/lib/python3.10/dist-packages (from matplotlib) (2.8.2)
Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.7->matplotlib) (1.16.0)
Requirement already satisfied: MarkupSafe>=2.0 in /usr/local/lib/python3.10/dist-packages (from jinja2->torch) (3.0.1)
Requirement already satisfied: mpmath<1.4,>=1.1.0 in /usr/local/lib/python3.10/dist-packages (from sympy->torch) (1.3.0)

import torch
import torch.nn.functional as F
from torchvision import models, transforms
from torchvision.models import ResNet18_Weights
from PIL import Image
import matplotlib.pyplot as plt
import numpy as np

# Load a pre-trained model
model = models.resnet18(weights=ResNet18_Weights.IMAGENET1K_V1)

# Lets look into the model
model.eval()

ResNet(
  (conv1): Conv2d(3, 64, kernel_size=(7, 7), stride=(2, 2), padding=(3, 3), bias=False)
  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (relu): ReLU(inplace=True)
  (maxpool): MaxPool2d(kernel_size=3, stride=2, padding=1, dilation=1, ceil_mode=False)
  (layer1): Sequential(
    (0): BasicBlock(
      (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    )
    (1): BasicBlock(
      (conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    )
  )
  (layer2): Sequential(
    (0): BasicBlock(
      (conv1): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (downsample): Sequential(
        (0): Conv2d(64, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): BasicBlock(
      (conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    )
  )
  (layer3): Sequential(
    (0): BasicBlock(
      (conv1): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (downsample): Sequential(
        (0): Conv2d(128, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): BasicBlock(
      (conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    )
  )
  (layer4): Sequential(
    (0): BasicBlock(
      (conv1): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (downsample): Sequential(
        (0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
        (1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      )
    )
    (1): BasicBlock(
      (conv1): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (relu): ReLU(inplace=True)
      (conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
      (bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    )
  )
  (avgpool): AdaptiveAvgPool2d(output_size=(1, 1))
  (fc): Linear(in_features=512, out_features=1000, bias=True)
)

19.4. ResNet-18 layer composition#

ResNet-18 has a total of 18 layers, which include convolutional layers and fully connected layers, structured as follows:

Input: Image size is typically (224 \times 224).
First Layer (Convolution + MaxPool):
- 7x7 convolution, 64 filters, stride 2, followed by a 3x3 max pooling layer.
Residual Blocks: These are the core building blocks. Each residual block consists of two convolutional layers followed by batch normalization and ReLU activation. ResNet-18 contains 8 of these residual blocks (4 stages with 2 residual blocks per stage):
1. Stage 1: 2 residual blocks, each with two 3x3 convolutions, 64 filters.
2. Stage 2: 2 residual blocks, each with two 3x3 convolutions, 128 filters.
3. Stage 3: 2 residual blocks, each with two 3x3 convolutions, 256 filters.
4. Stage 4: 2 residual blocks, each with two 3x3 convolutions, 512 filters.
In each of these stages, the first block increases the number of filters and reduces the spatial dimensions (when moving between stages), while the second block maintains the dimensions. When dimensions change, the identity connection (skip connection) is adjusted using a 1x1 convolution.
Fully Connected Layer:
- After the final residual block, the network applies global average pooling, followed by a fully connected (dense) layer with 1000 output units (for classification tasks like ImageNet).

19.5. Architecture Summary (Layer-by-Layer):#

Layer	Output Size	Details
Conv1	112x112	7x7, 64, stride 2
Max Pool	56x56	3x3 max pool, stride 2
Layer 1	56x56	2 x (3x3, 64)
Layer 2	28x28	2 x (3x3, 128), stride 2
Layer 3	14x14	2 x (3x3, 256), stride 2
Layer 4	7x7	2 x (3x3, 512), stride 2
Avg Pool	1x1	Global average pool
FC	1,000	Fully connected layer, 1,000 classes (ImageNet)

19.6. Residual Connections#

Residual connections and layers are fundamental components in the design of ResNet-18 (Residual Network with 18 layers), which is one of the architectures proposed to solve the problem of vanishing gradients and allow the training of much deeper networks. A residual connection, or a skip connection, is the hallmark feature of ResNet. The idea behind residual connections is to allow information to bypass one or more layers by adding the input directly to the output of a block of layers, creating a shortcut. This helps to alleviate the issue of vanishing or exploding gradients, enabling more effective training of deep networks.

A residual block typically looks like this:

Input (x) → Series of transformations (F(x)) → Output
Instead of only passing through the layers, ResNet adds a shortcut connection:

\[ \text{Output} = F(x) + x \]

where \(F(x)\) is the result of a sequence of operations like convolution, batch normalization, and ReLU activation. The identity connection (the term \(x\)) allows gradients to flow more easily during backpropagation, preventing them from becoming too small.

19.7. How Residual Connections Work in ResNet-18#

Each residual block in ResNet-18 has a structure like this:

Convolution 1 (3x3) → BatchNorm → ReLU → Convolution 2 (3x3) → BatchNorm.
The input to the block \(x\) is added to the output of the block \(F(x)\):

\[ \text{Output} = F(x) + x \]

Then, a ReLU activation is applied to the final output.

19.7.1. Key Benefits of Residual Connections:#

Ease of optimization: Residual connections help gradients flow through deeper networks, making optimization easier.
Addressing vanishing gradients: By adding skip connections, gradients have a clearer path during backpropagation, preventing them from becoming too small (vanishing gradients problem).
Training deeper networks: With residual connections, ResNet enables the training of very deep networks, such as ResNet-50, ResNet-101, and beyond, without suffering from significant degradation in performance.

In ResNet-18, this combination of residual blocks and skip connections allows the model to be trained efficiently, even with a depth of 18 layers, while still being less prone to overfitting and gradient-related issues.

19.8. Batch Normalization#

Batch normalization is a technique that normalizes the input to each layer in a neural network, improving the stability and speed of training. It was introduced to address the problem of internal covariate shift, where the distribution of inputs to a layer changes during training, causing slower convergence and making it harder to train deep networks.

Batch normalization works by normalizing the input of each mini-batch to have a mean of 0 and a standard deviation of 1. This ensures that the inputs to each layer have consistent distributions throughout the training process.

For an input (x) to a layer, batch normalization performs the following steps:

Compute mean and variance for the mini-batch:

\[ \mu_B = \frac{1}{m} \sum_{i=1}^{m} x_i \]

\[ \sigma_B^2 = \frac{1}{m} \sum_{i=1}^{m} (x_i - \mu_B)^2 \]

where \(m\) is the number of samples in the mini-batch, \(\mu_B\) is the batch mean, and \(\sigma_B^2\) is the batch variance.
Normalize each input by subtracting the mean and dividing by the standard deviation:

\[ \hat{x}_i = \frac{x_i - \mu_B}{\sqrt{\sigma_B^2 + \epsilon}} \]

where \(\epsilon\) is a small constant added to avoid division by zero.
Apply learnable scaling and shifting parameters \(\gamma\) and \(\beta\), which allow the network to learn the optimal scale and shift for each input:

\[ y_i = \gamma \hat{x}_i + \beta \]

These parameters enable the model to recover the original distribution of the inputs if necessary.

19.8.1. Benefits of Batch Normalization#

Batch normalization provides several key advantages:

Faster Training: By normalizing the inputs, the network can use higher learning rates without causing gradients to explode or vanish. This speeds up the convergence of the model.
Reduced Sensitivity to Initialization: Neural networks typically need careful weight initialization, but batch normalization reduces the sensitivity of the network to the initial parameters.
Regularization: Batch normalization adds a slight regularization effect because the noise introduced by mini-batch statistics during training acts as a form of stochastic noise, reducing the need for other regularization techniques like dropout.
Alleviates Internal Covariate Shift: By keeping the distribution of the inputs to each layer more stable during training, batch normalization helps the network converge more smoothly.

19.8.2. How Batch Normalization Works in Practice#

Batch normalization is typically applied after the linear transformation (convolution or fully connected layer) but before the non-linear activation function (like ReLU or Sigmoid). This sequence allows batch normalization to normalize the activations that will be fed into the non-linearity.

In a layer where batch normalization is applied, the order of operations is typically:

Convolution/Fully Connected Layer
Batch Normalization
Activation Function (ReLU, Sigmoid, etc.)

19.8.3. Batch Normalization During Training vs Inference#

During training: The mean and variance are computed for each mini-batch. The running averages of the mean and variance are maintained to be used later for inference.
During inference: Instead of using the statistics of the current batch, the running mean and variance (computed during training) are used to normalize the inputs. This ensures consistency between training and inference.

19.8.4. Example of Batch Normalization in ResNet-18#

In ResNet-18, batch normalization is applied after every convolution layer and before the ReLU activation. Each convolutional layer produces output feature maps, and batch normalization ensures that these feature maps have a stable distribution of values, improving training dynamics.

For example, in a residual block:

Convolution 1 → BatchNorm → ReLU → Convolution 2 → BatchNorm → Addition with Skip Connection → ReLU.

import requests
from PIL import Image
from io import BytesIO
import matplotlib.pyplot as plt

# URL of an example image from ImageNet (or you can replace it with any image URL)
image_url = "https://upload.wikimedia.org/wikipedia/commons/9/9a/Pug_600.jpg"  # Example: a pug image
image_url_cat = "https://user-images.githubusercontent.com/7392509/29206419-7a5eef52-7eb4-11e7-975c-98d0d7b6da99.JPG"
image_url_cat_dog = "https://media.istockphoto.com/id/1435010849/photo/labrador-retriever-dog-panting-and-ginger-cat-sitting-in-front-of-dark-yellow-background.jpg?s=612x612&w=0&k=20&c=obpeW_Aw7cIpQ8SfVHX9lMwLeLa00W1qmSFcOTVMEZ4="
image_url_boat_dog_human = "https://www.stockvault.net/data/2018/10/17/255259/preview16.jpg"


# define function to download and perform trasnsformation

def fetch_and_transform(url, transform, image = "image.jpg"):
    response = requests.get(url)
    img = Image.open(BytesIO(response.content))
    img.save(image)
    plt.imshow(img)
    plt.axis('off')
    plt.show()
    imgtensor = transform(img).unsqueeze(0)
    return img, imgtensor

# Preprocessing for the input image (as used in ImageNet)
transform = transforms.Compose([
    transforms.Resize((224, 224)),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])

pug_url = "https://upload.wikimedia.org/wikipedia/commons/9/9a/Pug_600.jpg"
pug_image, pug_tensor = fetch_and_transform(image_url, transform)

_images/587f5f5c17dba238a8918a2be6718c4f578a9e73d9a7c133a9fe1d37fa417958.png

pug_tensor.shape # 1 x channels x 224 x 224

torch.Size([1, 3, 224, 224])

import torch
import torch.nn.functional as F
from torchvision import models, transforms
from PIL import Image
import numpy as np
import matplotlib.pyplot as plt

def perform_forward(model, target_layer, image):
    """
    Perform a forward pass through the model and capture gradients and feature maps.
    """

    # Hook for gradients and feature maps
    gradients = []
    activations = []

    # Define a function to capture the Gradients
    def save_gradient(grad):
        gradients.append(grad)

    # Hook to capture the feature maps
    def forward_hook(module, input, output):
        activations.append(output)
        output.register_hook(save_gradient) # Hook to capture gradients

    hook = target_layer.register_forward_hook(forward_hook)

    output = model(image)
    # remove the hook
    hook.remove()

    return gradients, activations, output

# Register the hook to the last convolutional layer
target_layer = model.layer4[-1].conv2

gradients, activations, output = perform_forward(model, target_layer, pug_tensor)

# @title labels
labels = {0: 'tench, Tinca tinca',
'goldfish, Carassius auratus',
'great white shark, white shark, man-eater, man-eating shark, Carcharodon carcharias',
'tiger shark, Galeocerdo cuvieri',
'hammerhead, hammerhead shark',
'electric ray, crampfish, numbfish, torpedo',
'stingray',
'cock',
'hen',
'ostrich, Struthio camelus',
'brambling, Fringilla montifringilla',
'goldfinch, Carduelis carduelis',
'house finch, linnet, Carpodacus mexicanus',
'junco, snowbird',
'indigo bunting, indigo finch, indigo bird, Passerina cyanea',
'robin, American robin, Turdus migratorius',
'bulbul',
'jay',
'magpie',
'chickadee',
'water ouzel, dipper',
'kite',
'bald eagle, American eagle, Haliaeetus leucocephalus',
'vulture',
'great grey owl, great gray owl, Strix nebulosa',
'European fire salamander, Salamandra salamandra',
'common newt, Triturus vulgaris',
'eft',
'spotted salamander, Ambystoma maculatum',
'axolotl, mud puppy, Ambystoma mexicanum',
'bullfrog, Rana catesbeiana',
'tree frog, tree-frog',
'tailed frog, bell toad, ribbed toad, tailed toad, Ascaphus trui',
'loggerhead, loggerhead turtle, Caretta caretta',
'leatherback turtle, leatherback, leathery turtle, Dermochelys coriacea',
'mud turtle',
'terrapin',
'box turtle, box tortoise',
'banded gecko',
'common iguana, iguana, Iguana iguana',
'American chameleon, anole, Anolis carolinensis',
'whiptail, whiptail lizard',
'agama',
'frilled lizard, Chlamydosaurus kingi',
'alligator lizard',
'Gila monster, Heloderma suspectum',
'green lizard, Lacerta viridis',
'African chameleon, Chamaeleo chamaeleon',
'Komodo dragon, Komodo lizard, dragon lizard, giant lizard, Varanus komodoensis',
'African crocodile, Nile crocodile, Crocodylus niloticus',
'American alligator, Alligator mississipiensis',
'triceratops',
'thunder snake, worm snake, Carphophis amoenus',
'ringneck snake, ring-necked snake, ring snake',
'hognose snake, puff adder, sand viper',
'green snake, grass snake',
'king snake, kingsnake',
'garter snake, grass snake',
'water snake',
'vine snake',
'night snake, Hypsiglena torquata',
'boa constrictor, Constrictor constrictor',
'rock python, rock snake, Python sebae',
'Indian cobra, Naja naja',
'green mamba',
'sea snake',
'horned viper, cerastes, sand viper, horned asp, Cerastes cornutus',
'diamondback, diamondback rattlesnake, Crotalus adamanteus',
'sidewinder, horned rattlesnake, Crotalus cerastes',
'trilobite',
'harvestman, daddy longlegs, Phalangium opilio',
'scorpion',
'black and gold garden spider, Argiope aurantia',
'barn spider, Araneus cavaticus',
'garden spider, Aranea diademata',
'black widow, Latrodectus mactans',
'tarantula',
'wolf spider, hunting spider',
'tick',
'centipede',
'black grouse',
'ptarmigan',
'ruffed grouse, partridge, Bonasa umbellus',
'prairie chicken, prairie grouse, prairie fowl',
'peacock',
'quail',
'partridge',
'African grey, African gray, Psittacus erithacus',
'macaw',
'sulphur-crested cockatoo, Kakatoe galerita, Cacatua galerita',
'lorikeet',
'coucal',
'bee eater',
'hornbill',
'hummingbird',
'jacamar',
'toucan',
'drake',
'red-breasted merganser, Mergus serrator',
'goose',
'black swan, Cygnus atratus',
'tusker',
'echidna, spiny anteater, anteater',
'platypus, duckbill, duckbilled platypus, duck-billed platypus, Ornithorhynchus anatinus',
'wallaby, brush kangaroo',
'koala, koala bear, kangaroo bear, native bear, Phascolarctos cinereus',
'wombat',
'jellyfish',
'sea anemone, anemone',
'brain coral',
'flatworm, platyhelminth',
'nematode, nematode worm, roundworm',
'conch',
'snail',
'slug',
'sea slug, nudibranch',
'chiton, coat-of-mail shell, sea cradle, polyplacophore',
'chambered nautilus, pearly nautilus, nautilus',
'Dungeness crab, Cancer magister',
'rock crab, Cancer irroratus',
'fiddler crab',
'king crab, Alaska crab, Alaskan king crab, Alaska king crab, Paralithodes camtschatica',
'American lobster, Northern lobster, Maine lobster, Homarus americanus',
'spiny lobster, langouste, rock lobster, crawfish, crayfish, sea crawfish',
'crayfish, crawfish, crawdad, crawdaddy',
'hermit crab',
'isopod',
'white stork, Ciconia ciconia',
'black stork, Ciconia nigra',
'spoonbill',
'flamingo',
'little blue heron, Egretta caerulea',
'American egret, great white heron, Egretta albus',
'bittern',
'crane',
'limpkin, Aramus pictus',
'European gallinule, Porphyrio porphyrio',
'American coot, marsh hen, mud hen, water hen, Fulica americana',
'bustard',
'ruddy turnstone, Arenaria interpres',
'red-backed sandpiper, dunlin, Erolia alpina',
'redshank, Tringa totanus',
'dowitcher',
'oystercatcher, oyster catcher',
'pelican',
'king penguin, Aptenodytes patagonica',
'albatross, mollymawk',
'grey whale, gray whale, devilfish, Eschrichtius gibbosus, Eschrichtius robustus',
'killer whale, killer, orca, grampus, sea wolf, Orcinus orca',
'dugong, Dugong dugon',
'sea lion',
'Chihuahua',
'Japanese spaniel',
'Maltese dog, Maltese terrier, Maltese',
'Pekinese, Pekingese, Peke',
'Shih-Tzu',
'Blenheim spaniel',
'papillon',
'toy terrier',
'Rhodesian ridgeback',
'Afghan hound, Afghan',
'basset, basset hound',
'beagle',
'bloodhound, sleuthhound',
'bluetick',
'black-and-tan coonhound',
'Walker hound, Walker foxhound',
'English foxhound',
'redbone',
'borzoi, Russian wolfhound',
'Irish wolfhound',
'Italian greyhound',
'whippet',
'Ibizan hound, Ibizan Podenco',
'Norwegian elkhound, elkhound',
'otterhound, otter hound',
'Saluki, gazelle hound',
'Scottish deerhound, deerhound',
'Weimaraner',
'Staffordshire bullterrier, Staffordshire bull terrier',
'American Staffordshire terrier, Staffordshire terrier, American pit bull terrier, pit bull terrier',
'Bedlington terrier',
'Border terrier',
'Kerry blue terrier',
'Irish terrier',
'Norfolk terrier',
'Norwich terrier',
'Yorkshire terrier',
'wire-haired fox terrier',
'Lakeland terrier',
'Sealyham terrier, Sealyham',
'Airedale, Airedale terrier',
'cairn, cairn terrier',
'Australian terrier',
'Dandie Dinmont, Dandie Dinmont terrier',
'Boston bull, Boston terrier',
'miniature schnauzer',
'giant schnauzer',
'standard schnauzer',
'Scotch terrier, Scottish terrier, Scottie',
'Tibetan terrier, chrysanthemum dog',
'silky terrier, Sydney silky',
'soft-coated wheaten terrier',
'West Highland white terrier',
'Lhasa, Lhasa apso',
'flat-coated retriever',
'curly-coated retriever',
'golden retriever',
'Labrador retriever',
'Chesapeake Bay retriever',
'German short-haired pointer',
'vizsla, Hungarian pointer',
'English setter',
'Irish setter, red setter',
'Gordon setter',
'Brittany spaniel',
'clumber, clumber spaniel',
'English springer, English springer spaniel',
'Welsh springer spaniel',
'cocker spaniel, English cocker spaniel, cocker',
'Sussex spaniel',
'Irish water spaniel',
'kuvasz',
'schipperke',
'groenendael',
'malinois',
'briard',
'kelpie',
'komondor',
'Old English sheepdog, bobtail',
'Shetland sheepdog, Shetland sheep dog, Shetland',
'collie',
'Border collie',
'Bouvier des Flandres, Bouviers des Flandres',
'Rottweiler',
'German shepherd, German shepherd dog, German police dog, alsatian',
'Doberman, Doberman pinscher',
'miniature pinscher',
'Greater Swiss Mountain dog',
'Bernese mountain dog',
'Appenzeller',
'EntleBucher',
'boxer',
'bull mastiff',
'Tibetan mastiff',
'French bulldog',
'Great Dane',
'Saint Bernard, St Bernard',
'Eskimo dog, husky',
'malamute, malemute, Alaskan malamute',
'Siberian husky',
'dalmatian, coach dog, carriage dog',
'affenpinscher, monkey pinscher, monkey dog',
'basenji',
'pug, pug-dog',
'Leonberg',
'Newfoundland, Newfoundland dog',
'Great Pyrenees',
'Samoyed, Samoyede',
'Pomeranian',
'chow, chow chow',
'keeshond',
'Brabancon griffon',
'Pembroke, Pembroke Welsh corgi',
'Cardigan, Cardigan Welsh corgi',
'toy poodle',
'miniature poodle',
'standard poodle',
'Mexican hairless',
'timber wolf, grey wolf, gray wolf, Canis lupus',
'white wolf, Arctic wolf, Canis lupus tundrarum',
'red wolf, maned wolf, Canis rufus, Canis niger',
'coyote, prairie wolf, brush wolf, Canis latrans',
'dingo, warrigal, warragal, Canis dingo',
'dhole, Cuon alpinus',
'African hunting dog, hyena dog, Cape hunting dog, Lycaon pictus',
'hyena, hyaena',
'red fox, Vulpes vulpes',
'kit fox, Vulpes macrotis',
'Arctic fox, white fox, Alopex lagopus',
'grey fox, gray fox, Urocyon cinereoargenteus',
'tabby, tabby cat',
'tiger cat',
'Persian cat',
'Siamese cat, Siamese',
'Egyptian cat',
'cougar, puma, catamount, mountain lion, painter, panther, Felis concolor',
'lynx, catamount',
'leopard, Panthera pardus',
'snow leopard, ounce, Panthera uncia',
'jaguar, panther, Panthera onca, Felis onca',
'lion, king of beasts, Panthera leo',
'tiger, Panthera tigris',
'cheetah, chetah, Acinonyx jubatus',
'brown bear, bruin, Ursus arctos',
'American black bear, black bear, Ursus americanus, Euarctos americanus',
'ice bear, polar bear, Ursus Maritimus, Thalarctos maritimus',
'sloth bear, Melursus ursinus, Ursus ursinus',
'mongoose',
'meerkat, mierkat',
'tiger beetle',
'ladybug, ladybeetle, lady beetle, ladybird, ladybird beetle',
'ground beetle, carabid beetle',
'long-horned beetle, longicorn, longicorn beetle',
'leaf beetle, chrysomelid',
'dung beetle',
'rhinoceros beetle',
'weevil',
'fly',
'bee',
'ant, emmet, pismire',
'grasshopper, hopper',
'cricket',
'walking stick, walkingstick, stick insect',
'cockroach, roach',
'mantis, mantid',
'cicada, cicala',
'leafhopper',
'lacewing, lacewing fly',
"dragonfly, darning needle, devil's darning needle, sewing needle, snake feeder, snake doctor, mosquito hawk, skeeter hawk",
'damselfly',
'admiral',
'ringlet, ringlet butterfly',
'monarch, monarch butterfly, milkweed butterfly, Danaus plexippus',
'cabbage butterfly',
'sulphur butterfly, sulfur butterfly',
'lycaenid, lycaenid butterfly',
'starfish, sea star',
'sea urchin',
'sea cucumber, holothurian',
'wood rabbit, cottontail, cottontail rabbit',
'hare',
'Angora, Angora rabbit',
'hamster',
'porcupine, hedgehog',
'fox squirrel, eastern fox squirrel, Sciurus niger',
'marmot',
'beaver',
'guinea pig, Cavia cobaya',
'sorrel',
'zebra',
'hog, pig, grunter, squealer, Sus scrofa',
'wild boar, boar, Sus scrofa',
'warthog',
'hippopotamus, hippo, river horse, Hippopotamus amphibius',
'ox',
'water buffalo, water ox, Asiatic buffalo, Bubalus bubalis',
'bison',
'ram, tup',
'bighorn, bighorn sheep, cimarron, Rocky Mountain bighorn, Rocky Mountain sheep, Ovis canadensis',
'ibex, Capra ibex',
'hartebeest',
'impala, Aepyceros melampus',
'gazelle',
'Arabian camel, dromedary, Camelus dromedarius',
'llama',
'weasel',
'mink',
'polecat, fitch, foulmart, foumart, Mustela putorius',
'black-footed ferret, ferret, Mustela nigripes',
'otter',
'skunk, polecat, wood pussy',
'badger',
'armadillo',
'three-toed sloth, ai, Bradypus tridactylus',
'orangutan, orang, orangutang, Pongo pygmaeus',
'gorilla, Gorilla gorilla',
'chimpanzee, chimp, Pan troglodytes',
'gibbon, Hylobates lar',
'siamang, Hylobates syndactylus, Symphalangus syndactylus',
'guenon, guenon monkey',
'patas, hussar monkey, Erythrocebus patas',
'baboon',
'macaque',
'langur',
'colobus, colobus monkey',
'proboscis monkey, Nasalis larvatus',
'marmoset',
'capuchin, ringtail, Cebus capucinus',
'howler monkey, howler',
'titi, titi monkey',
'spider monkey, Ateles geoffroyi',
'squirrel monkey, Saimiri sciureus',
'Madagascar cat, ring-tailed lemur, Lemur catta',
'indri, indris, Indri indri, Indri brevicaudatus',
'Indian elephant, Elephas maximus',
'African elephant, Loxodonta africana',
'lesser panda, red panda, panda, bear cat, cat bear, Ailurus fulgens',
'giant panda, panda, panda bear, coon bear, Ailuropoda melanoleuca',
'barracouta, snoek',
'eel',
'coho, cohoe, coho salmon, blue jack, silver salmon, Oncorhynchus kisutch',
'rock beauty, Holocanthus tricolor',
'anemone fish',
'sturgeon',
'gar, garfish, garpike, billfish, Lepisosteus osseus',
'lionfish',
'puffer, pufferfish, blowfish, globefish',
'abacus',
'abaya',
"academic gown, academic robe, judge's robe",
'accordion, piano accordion, squeeze box',
'acoustic guitar',
'aircraft carrier, carrier, flattop, attack aircraft carrier',
'airliner',
'airship, dirigible',
'altar',
'ambulance',
'amphibian, amphibious vehicle',
'analog clock',
'apiary, bee house',
'apron',
'ashcan, trash can, garbage can, wastebin, ash bin, ash-bin, ashbin, dustbin, trash barrel, trash bin',
'assault rifle, assault gun',
'backpack, back pack, knapsack, packsack, rucksack, haversack',
'bakery, bakeshop, bakehouse',
'balance beam, beam',
'balloon',
'ballpoint, ballpoint pen, ballpen, Biro',
'Band Aid',
'banjo',
'bannister, banister, balustrade, balusters, handrail',
'barbell',
'barber chair',
'barbershop',
'barn',
'barometer',
'barrel, cask',
'barrow, garden cart, lawn cart, wheelbarrow',
'baseball',
'basketball',
'bassinet',
'bassoon',
'bathing cap, swimming cap',
'bath towel',
'bathtub, bathing tub, bath, tub',
'beach wagon, station wagon, wagon, estate car, beach waggon, station waggon, waggon',
'beacon, lighthouse, beacon light, pharos',
'beaker',
'bearskin, busby, shako',
'beer bottle',
'beer glass',
'bell cote, bell cot',
'bib',
'bicycle-built-for-two, tandem bicycle, tandem',
'bikini, two-piece',
'binder, ring-binder',
'binoculars, field glasses, opera glasses',
'birdhouse',
'boathouse',
'bobsled, bobsleigh, bob',
'bolo tie, bolo, bola tie, bola',
'bonnet, poke bonnet',
'bookcase',
'bookshop, bookstore, bookstall',
'bottlecap',
'bow',
'bow tie, bow-tie, bowtie',
'brass, memorial tablet, plaque',
'brassiere, bra, bandeau',
'breakwater, groin, groyne, mole, bulwark, seawall, jetty',
'breastplate, aegis, egis',
'broom',
'bucket, pail',
'buckle',
'bulletproof vest',
'bullet train, bullet',
'butcher shop, meat market',
'cab, hack, taxi, taxicab',
'caldron, cauldron',
'candle, taper, wax light',
'cannon',
'canoe',
'can opener, tin opener',
'cardigan',
'car mirror',
'carousel, carrousel, merry-go-round, roundabout, whirligig',
"carpenter's kit, tool kit",
'carton',
'car wheel',
'cash machine, cash dispenser, automated teller machine, automatic teller machine, automated teller, automatic teller, ATM',
'cassette',
'cassette player',
'castle',
'catamaran',
'CD player',
'cello, violoncello',
'cellular telephone, cellular phone, cellphone, cell, mobile phone',
'chain',
'chainlink fence',
'chain mail, ring mail, mail, chain armor, chain armour, ring armor, ring armour',
'chain saw, chainsaw',
'chest',
'chiffonier, commode',
'chime, bell, gong',
'china cabinet, china closet',
'Christmas stocking',
'church, church building',
'cinema, movie theater, movie theatre, movie house, picture palace',
'cleaver, meat cleaver, chopper',
'cliff dwelling',
'cloak',
'clog, geta, patten, sabot',
'cocktail shaker',
'coffee mug',
'coffeepot',
'coil, spiral, volute, whorl, helix',
'combination lock',
'computer keyboard, keypad',
'confectionery, confectionary, candy store',
'container ship, containership, container vessel',
'convertible',
'corkscrew, bottle screw',
'cornet, horn, trumpet, trump',
'cowboy boot',
'cowboy hat, ten-gallon hat',
'cradle',
'crane',
'crash helmet',
'crate',
'crib, cot',
'Crock Pot',
'croquet ball',
'crutch',
'cuirass',
'dam, dike, dyke',
'desk',
'desktop computer',
'dial telephone, dial phone',
'diaper, nappy, napkin',
'digital clock',
'digital watch',
'dining table, board',
'dishrag, dishcloth',
'dishwasher, dish washer, dishwashing machine',
'disk brake, disc brake',
'dock, dockage, docking facility',
'dogsled, dog sled, dog sleigh',
'dome',
'doormat, welcome mat',
'drilling platform, offshore rig',
'drum, membranophone, tympan',
'drumstick',
'dumbbell',
'Dutch oven',
'electric fan, blower',
'electric guitar',
'electric locomotive',
'entertainment center',
'envelope',
'espresso maker',
'face powder',
'feather boa, boa',
'file, file cabinet, filing cabinet',
'fireboat',
'fire engine, fire truck',
'fire screen, fireguard',
'flagpole, flagstaff',
'flute, transverse flute',
'folding chair',
'football helmet',
'forklift',
'fountain',
'fountain pen',
'four-poster',
'freight car',
'French horn, horn',
'frying pan, frypan, skillet',
'fur coat',
'garbage truck, dustcart',
'gasmask, respirator, gas helmet',
'gas pump, gasoline pump, petrol pump, island dispenser',
'goblet',
'go-kart',
'golf ball',
'golfcart, golf cart',
'gondola',
'gong, tam-tam',
'gown',
'grand piano, grand',
'greenhouse, nursery, glasshouse',
'grille, radiator grille',
'grocery store, grocery, food market, market',
'guillotine',
'hair slide',
'hair spray',
'half track',
'hammer',
'hamper',
'hand blower, blow dryer, blow drier, hair dryer, hair drier',
'hand-held computer, hand-held microcomputer',
'handkerchief, hankie, hanky, hankey',
'hard disc, hard disk, fixed disk',
'harmonica, mouth organ, harp, mouth harp',
'harp',
'harvester, reaper',
'hatchet',
'holster',
'home theater, home theatre',
'honeycomb',
'hook, claw',
'hoopskirt, crinoline',
'horizontal bar, high bar',
'horse cart, horse-cart',
'hourglass',
'iPod',
'iron, smoothing iron',
"jack-o'-lantern",
'jean, blue jean, denim',
'jeep, landrover',
'jersey, T-shirt, tee shirt',
'jigsaw puzzle',
'jinrikisha, ricksha, rickshaw',
'joystick',
'kimono',
'knee pad',
'knot',
'lab coat, laboratory coat',
'ladle',
'lampshade, lamp shade',
'laptop, laptop computer',
'lawn mower, mower',
'lens cap, lens cover',
'letter opener, paper knife, paperknife',
'library',
'lifeboat',
'lighter, light, igniter, ignitor',
'limousine, limo',
'liner, ocean liner',
'lipstick, lip rouge',
'Loafer',
'lotion',
'loudspeaker, speaker, speaker unit, loudspeaker system, speaker system',
"loupe, jeweler's loupe",
'lumbermill, sawmill',
'magnetic compass',
'mailbag, postbag',
'mailbox, letter box',
'maillot',
'maillot, tank suit',
'manhole cover',
'maraca',
'marimba, xylophone',
'mask',
'matchstick',
'maypole',
'maze, labyrinth',
'measuring cup',
'medicine chest, medicine cabinet',
'megalith, megalithic structure',
'microphone, mike',
'microwave, microwave oven',
'military uniform',
'milk can',
'minibus',
'miniskirt, mini',
'minivan',
'missile',
'mitten',
'mixing bowl',
'mobile home, manufactured home',
'Model T',
'modem',
'monastery',
'monitor',
'moped',
'mortar',
'mortarboard',
'mosque',
'mosquito net',
'motor scooter, scooter',
'mountain bike, all-terrain bike, off-roader',
'mountain tent',
'mouse, computer mouse',
'mousetrap',
'moving van',
'muzzle',
'nail',
'neck brace',
'necklace',
'nipple',
'notebook, notebook computer',
'obelisk',
'oboe, hautboy, hautbois',
'ocarina, sweet potato',
'odometer, hodometer, mileometer, milometer',
'oil filter',
'organ, pipe organ',
'oscilloscope, scope, cathode-ray oscilloscope, CRO',
'overskirt',
'oxcart',
'oxygen mask',
'packet',
'paddle, boat paddle',
'paddlewheel, paddle wheel',
'padlock',
'paintbrush',
"pajama, pyjama, pj's, jammies",
'palace',
'panpipe, pandean pipe, syrinx',
'paper towel',
'parachute, chute',
'parallel bars, bars',
'park bench',
'parking meter',
'passenger car, coach, carriage',
'patio, terrace',
'pay-phone, pay-station',
'pedestal, plinth, footstall',
'pencil box, pencil case',
'pencil sharpener',
'perfume, essence',
'Petri dish',
'photocopier',
'pick, plectrum, plectron',
'pickelhaube',
'picket fence, paling',
'pickup, pickup truck',
'pier',
'piggy bank, penny bank',
'pill bottle',
'pillow',
'ping-pong ball',
'pinwheel',
'pirate, pirate ship',
'pitcher, ewer',
"plane, carpenter's plane, woodworking plane",
'planetarium',
'plastic bag',
'plate rack',
'plow, plough',
"plunger, plumber's helper",
'Polaroid camera, Polaroid Land camera',
'pole',
'police van, police wagon, paddy wagon, patrol wagon, wagon, black Maria',
'poncho',
'pool table, billiard table, snooker table',
'pop bottle, soda bottle',
'pot, flowerpot',
"potter's wheel",
'power drill',
'prayer rug, prayer mat',
'printer',
'prison, prison house',
'projectile, missile',
'projector',
'puck, hockey puck',
'punching bag, punch bag, punching ball, punchball',
'purse',
'quill, quill pen',
'quilt, comforter, comfort, puff',
'racer, race car, racing car',
'racket, racquet',
'radiator',
'radio, wireless',
'radio telescope, radio reflector',
'rain barrel',
'recreational vehicle, RV, R.V.',
'reel',
'reflex camera',
'refrigerator, icebox',
'remote control, remote',
'restaurant, eating house, eating place, eatery',
'revolver, six-gun, six-shooter',
'rifle',
'rocking chair, rocker',
'rotisserie',
'rubber eraser, rubber, pencil eraser',
'rugby ball',
'rule, ruler',
'running shoe',
'safe',
'safety pin',
'saltshaker, salt shaker',
'sandal',
'sarong',
'sax, saxophone',
'scabbard',
'scale, weighing machine',
'school bus',
'schooner',
'scoreboard',
'screen, CRT screen',
'screw',
'screwdriver',
'seat belt, seatbelt',
'sewing machine',
'shield, buckler',
'shoe shop, shoe-shop, shoe store',
'shoji',
'shopping basket',
'shopping cart',
'shovel',
'shower cap',
'shower curtain',
'ski',
'ski mask',
'sleeping bag',
'slide rule, slipstick',
'sliding door',
'slot, one-armed bandit',
'snorkel',
'snowmobile',
'snowplow, snowplough',
'soap dispenser',
'soccer ball',
'sock',
'solar dish, solar collector, solar furnace',
'sombrero',
'soup bowl',
'space bar',
'space heater',
'space shuttle',
'spatula',
'speedboat',
"spider web, spider's web",
'spindle',
'sports car, sport car',
'spotlight, spot',
'stage',
'steam locomotive',
'steel arch bridge',
'steel drum',
'stethoscope',
'stole',
'stone wall',
'stopwatch, stop watch',
'stove',
'strainer',
'streetcar, tram, tramcar, trolley, trolley car',
'stretcher',
'studio couch, day bed',
'stupa, tope',
'submarine, pigboat, sub, U-boat',
'suit, suit of clothes',
'sundial',
'sunglass',
'sunglasses, dark glasses, shades',
'sunscreen, sunblock, sun blocker',
'suspension bridge',
'swab, swob, mop',
'sweatshirt',
'swimming trunks, bathing trunks',
'swing',
'switch, electric switch, electrical switch',
'syringe',
'table lamp',
'tank, army tank, armored combat vehicle, armoured combat vehicle',
'tape player',
'teapot',
'teddy, teddy bear',
'television, television system',
'tennis ball',
'thatch, thatched roof',
'theater curtain, theatre curtain',
'thimble',
'thresher, thrasher, threshing machine',
'throne',
'tile roof',
'toaster',
'tobacco shop, tobacconist shop, tobacconist',
'toilet seat',
'torch',
'totem pole',
'tow truck, tow car, wrecker',
'toyshop',
'tractor',
'trailer truck, tractor trailer, trucking rig, rig, articulated lorry, semi',
'tray',
'trench coat',
'tricycle, trike, velocipede',
'trimaran',
'tripod',
'triumphal arch',
'trolleybus, trolley coach, trackless trolley',
'trombone',
'tub, vat',
'turnstile',
'typewriter keyboard',
'umbrella',
'unicycle, monocycle',
'upright, upright piano',
'vacuum, vacuum cleaner',
'vase',
'vault',
'velvet',
'vending machine',
'vestment',
'viaduct',
'violin, fiddle',
'volleyball',
'waffle iron',
'wall clock',
'wallet, billfold, notecase, pocketbook',
'wardrobe, closet, press',
'warplane, military plane',
'washbasin, handbasin, washbowl, lavabo, wash-hand basin',
'washer, automatic washer, washing machine',
'water bottle',
'water jug',
'water tower',
'whiskey jug',
'whistle',
'wig',
'window screen',
'window shade',
'Windsor tie',
'wine bottle',
'wing',
'wok',
'wooden spoon',
'wool, woolen, woollen',
'worm fence, snake fence, snake-rail fence, Virginia fence',
'wreck',
'yawl',
'yurt',
'web site, website, internet site, site',
'comic book',
'crossword puzzle, crossword',
'street sign',
'traffic light, traffic signal, stoplight',
'book jacket, dust cover, dust jacket, dust wrapper',
'menu',
'plate',
'guacamole',
'consomme',
'hot pot, hotpot',
'trifle',
'ice cream, icecream',
'ice lolly, lolly, lollipop, popsicle',
'French loaf',
'bagel, beigel',
'pretzel',
'cheeseburger',
'hotdog, hot dog, red hot',
'mashed potato',
'head cabbage',
'broccoli',
'cauliflower',
'zucchini, courgette',
'spaghetti squash',
'acorn squash',
'butternut squash',
'cucumber, cuke',
'artichoke, globe artichoke',
'bell pepper',
'cardoon',
'mushroom',
'Granny Smith',
'strawberry',
'orange',
'lemon',
'fig',
'pineapple, ananas',
'banana',
'jackfruit, jak, jack',
'custard apple',
'pomegranate',
'hay',
'carbonara',
'chocolate sauce, chocolate syrup',
'dough',
'meat loaf, meatloaf',
'pizza, pizza pie',
'potpie',
'burrito',
'red wine',
'espresso',
'cup',
'eggnog',
'alp',
'bubble',
'cliff, drop, drop-off',
'coral reef',
'geyser',
'lakeside, lakeshore',
'promontory, headland, head, foreland',
'sandbar, sand bar',
'seashore, coast, seacoast, sea-coast',
'valley, vale',
'volcano',
'ballplayer, baseball player',
'groom, bridegroom',
'scuba diver',
'rapeseed',
'daisy',
"yellow lady's slipper, yellow lady-slipper, Cypripedium calceolus, Cypripedium parviflorum",
'corn',
'acorn',
'hip, rose hip, rosehip',
'buckeye, horse chestnut, conker',
'coral fungus',
'agaric',
'gyromitra',
'stinkhorn, carrion fungus',
'earthstar',
'hen-of-the-woods, hen of the woods, Polyporus frondosus, Grifola frondosa',
'bolete',
'ear, spike, capitulum',
'toilet tissue, toilet paper, bathroom tissue'
 }

reverse_labels = {v: k for k, v in labels.items()}

# lets print the best 10 results.
print (*[labels[i] for i in
         output.argsort(descending = True).tolist()[0][:10]
         ],
       sep = "\n"
)

pug, pug-dog
bull mastiff
French bulldog
Brabancon griffon
Saint Bernard, St Bernard
Norwegian elkhound, elkhound
Pekinese, Pekingese, Peke
chow, chow chow
Lhasa, Lhasa apso
Shih-Tzu

# lets take the max prediction
pred_class = output.argmax().item()

# Backward pass to get gradients of the predicted class
model.zero_grad()
# Set all others except pred_class to 0
for i in range(len(output[0])):
    if i != pred_class:
        output[0, i] = 0
print (output[0, pred_class])
output[0, pred_class].backward()

# Now the gradients and feature maps are saved in the lists `gradients` and `activations`

tensor(20.6269, grad_fn=<SelectBackward0>)

# lets check whatis the size of the gradients and activations
print (len(gradients))
print (len(activations))

4
1

print (gradients[0].shape)
print (activations[0].shape)

torch.Size([1, 512, 7, 7])
torch.Size([1, 512, 7, 7])

# Get the gradients and feature maps

# gradients w.r.t c^{th} label
gradient = gradients[0]  # Gradients captured during backward pass

# Get all the k activations
feature_maps = activations[0]  # Feature maps captured during forward pass

# sum over the spatial dimensions i, j in the k activations.
weights = torch.mean(gradient, dim=[2, 3], keepdim=True)  # Average pooling over spatial dimensions

# Weighted sum of feature maps to get the Grad-CAM heatmap Perform weighted sum of activations
cam = torch.sum(weights * feature_maps, dim=1).squeeze().detach().cpu().numpy()
print ("cam shap eis :", cam.shape)

# Apply ReLU and normalize the cam
cam = np.maximum(cam, 0)  # ReLU
cam = cam / cam.max()  # Normalize to [0, 1]

import matplotlib.pyplot as plt

plt.imshow(cam, cmap='jet')
plt.axis('off')
plt.show()

cam shap eis : (7, 7)

_images/2c58109f96342822f92e23ab8abca9830157dbe5e010cd7140b23bbd53cd510b.png

# Rescale heatmap to match the input image size
import cv2
cam = cv2.resize(cam, (pug_tensor.shape[2], pug_tensor.shape[3]))
in_image = cv2.resize(np.array(pug_image), (pug_tensor.shape[2], pug_tensor.shape[3]))

# Visualize the heatmap
plt.imshow(in_image)
plt.imshow(cam, cmap='jet', alpha=0.5)  # Overlay heatmap with transparency
plt.show()

_images/3e616c83d1ee58218925e29da816bf363357a1046996cd08346b9d772d4dec48.png

19.9. Lets try with a more complex image#

image_url_cat_dog = "https://media.istockphoto.com/id/1435010849/photo/labrador-retriever-dog-panting-and-ginger-cat-sitting-in-front-of-dark-yellow-background.jpg?s=612x612&w=0&k=20&c=obpeW_Aw7cIpQ8SfVHX9lMwLeLa00W1qmSFcOTVMEZ4="
image_url_boat_dog_human = "https://www.stockvault.net/data/2018/10/17/255259/preview16.jpg"

cat_dog_image, cat_dog_tensor = fetch_and_transform(image_url_cat_dog, transform)

_images/644ecbb519e90f5524e73dbb7a40c8526e201349a680465aad7f367942fc07b4.png

target_layer = model.layer4[-1].conv2
gradients, activations, output = perform_forward(model, target_layer, cat_dog_tensor)

# lets print the best 10 results.
print (*[labels[i] for i in
         output.argsort(descending = True).tolist()[0][:10]
         ],
       sep = "\n"
)

Labrador retriever
golden retriever
Rhodesian ridgeback
Chesapeake Bay retriever
tennis ball
kuvasz
redbone
English foxhound
bull mastiff
bloodhound, sleuthhound

cat = 281 # also, 282 283 284 285 286 287 are all cats
dog = 208 # for labrador retriever


def grad_cam(model, label, image, target_layer):
    target_layer = model.layer4[-1].conv2
    gradients, activations, output = perform_forward(model, target_layer, image)
    pred_class = label
    # Backward pass to get gradients of the predicted class
    model.zero_grad()
    # Set all others except pred_class to 0
    for i in range(len(output[0])):
        if i != pred_class:
            output[0, i] = 0
    print (output[0, pred_class])
    output[0, pred_class].backward()

    # Get the gradients and feature maps

    # gradients w.r.t c^{th} label
    gradient = gradients[0]  # Gradients captured during backward pass

    # Get all the k activations
    feature_maps = activations[0]  # Feature maps captured during forward pass

    # sum over the spatial dimensions i, j in the k activations.
    weights = torch.mean(gradient, dim=[2, 3], keepdim=True)  # Average pooling over spatial dimensions

    # Weighted sum of feature maps to get the Grad-CAM heatmap Perform weighted sum of activations
    cam = torch.sum(weights * feature_maps, dim=1).squeeze().detach().cpu().numpy()
    print ("cam shap eis :", cam.shape)

    # Apply ReLU and normalize the cam
    cam = np.maximum(cam, 0)  # ReLU
    cam = cam / cam.max()  # Normalize to [0, 1]

    return cam


dog_grad_cam = grad_cam(model, dog, cat_dog_tensor, target_layer)
cat_grad_cam = grad_cam(model, cat, cat_dog_tensor, target_layer)

# Rescale heatmap to match the input image size
import cv2
dog_grad_cam = cv2.resize(dog_grad_cam, (cat_dog_tensor.shape[2], cat_dog_tensor.shape[3]))
cat_grad_cam = cv2.resize(cat_grad_cam, (cat_dog_tensor.shape[2], cat_dog_tensor.shape[3]))
in_image = cv2.resize(np.array(cat_dog_image), (cat_dog_tensor.shape[2], cat_dog_tensor.shape[3]))

# make subplots with 2 columns fo rthe two grads

fig, axes = plt.subplots(1, 2, figsize=(10, 5))

# plot the cat grad cam

axes[0].imshow(in_image)
axes[0].imshow(cat_grad_cam, cmap='jet', alpha=0.5)  # Overlay heatmap with transparency
axes[0].set_title('Cat Grad-CAM')

axes[1].imshow(in_image)
axes[1].imshow(dog_grad_cam, cmap='jet', alpha=0.5)  # Overlay heatmap with transparency
axes[1].set_title('Dog Grad-CAM')

plt.show()

tensor(10.3447, grad_fn=<SelectBackward0>)
cam shap eis : (7, 7)
tensor(2.4705, grad_fn=<SelectBackward0>)
cam shap eis : (7, 7)

_images/5ca3630fea32a8e561d7d75425199c0bb4a102c9d840da63d69e32a5239355d4.png