Note
Go to the end to download the full example code
Image Classification Tutorial#
In this tutorial, you will learn how to validate your classification model using deepchecks test suites. You can read more about the different checks and suites for computer vision use cases at the examples section.
A classification model is usually used to classify an image into one of a number of classes. Although there are multi label use-cases, in which the model is used to classify an image into multiple classes, most use-cases require the model to classify images into a single class. Currently, deepchecks supports only single label classification (either binary or multi-class).
# Before we start, if you don't have deepchecks vision package installed yet, run:
import sys
!{sys.executable} -m pip install "deepchecks[vision]" --quiet --upgrade # --user
# or install using pip from your python environment
Defining the data and model#
Note
In this tutorial, we use the pytorch to create the dataset and model. To see how this can be done using tensorflow or other frameworks, please visit the creating VisionData guide.
Downloading the dataset#
The data is available from the torch library. We will download and extract it to the current directory.
import os
import urllib.request
import zipfile
url = 'https://download.pytorch.org/tutorial/hymenoptera_data.zip'
urllib.request.urlretrieve(url, './hymenoptera_data.zip')
with zipfile.ZipFile('./hymenoptera_data.zip', 'r') as zip_ref:
zip_ref.extractall('.')
Load Data#
We will use torchvision and torch.utils.data packages for loading the data. The model we are building will learn to classify ants and bees. We have about 120 training images each for ants and bees. There are 75 validation images for each class. This dataset is a very small subset of imagenet.
import albumentations as A
import numpy as np
import PIL.Image
import torch
import torchvision
from albumentations.pytorch import ToTensorV2
from torch import nn
from torch.utils.data import DataLoader
class AntsBeesDataset(torchvision.datasets.ImageFolder):
def __getitem__(self, index: int):
"""overrides __getitem__ to be compatible to albumentations"""
path, target = self.samples[index]
sample = self.loader(path)
sample = self.get_cv2_image(sample)
if self.transforms is not None:
transformed = self.transforms(image=sample, target=target)
sample, target = transformed["image"], transformed["target"]
else:
if self.transform is not None:
sample = self.transform(image=sample)['image']
if self.target_transform is not None:
target = self.target_transform(target)
return sample, target
def get_cv2_image(self, image):
if isinstance(image, PIL.Image.Image):
return np.array(image).astype('uint8')
elif isinstance(image, np.ndarray):
return image
else:
raise RuntimeError("Only PIL.Image and CV2 loaders currently supported!")
data_dir = './hymenoptera_data'
# Just normalization for validation
data_transforms = A.Compose([
A.Resize(height=256, width=256),
A.CenterCrop(height=224, width=224),
A.Normalize(mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225)),
ToTensorV2(),
])
train_dataset = AntsBeesDataset(root=os.path.join(data_dir,'train'))
train_dataset.transforms = data_transforms
test_dataset = AntsBeesDataset(root=os.path.join(data_dir, 'val'))
test_dataset.transforms = data_transforms
Visualize the dataset#
Let’s see how our data looks like.
print(f'Number of training images: {len(train_dataset)}')
print(f'Number of validation images: {len(test_dataset)}')
print(f'Example output of an image shape: {train_dataset[0][0].shape}')
print(f'Example output of a label: {train_dataset[0][1]}')
Number of training images: 244
Number of validation images: 153
Example output of an image shape: torch.Size([3, 224, 224])
Example output of a label: 0
Downloading a pre-trained model#
Now, we will download a pre-trained model from torchvision, that was trained on the ImageNet dataset.
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
model = torchvision.models.resnet18(pretrained=True)
num_ftrs = model.fc.in_features
# We have only 2 classes
model.fc = nn.Linear(num_ftrs, 2)
model = model.to(device)
_ = model.eval()
Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /home/runner/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
0%| | 0.00/44.7M [00:00<?, ?B/s]
47%|####6 | 20.9M/44.7M [00:00<00:00, 220MB/s]
100%|##########| 44.7M/44.7M [00:00<00:00, 264MB/s]
Validating the Model with Deepchecks#
Now, after we have the training data, validation data and the model, we can validate the model with deepchecks test suites.
Implementing the VisionData class#
The checks in the package validate the model & data by calculating various quantities over the data, labels and
predictions. In order to do that, those must be in a pre-defined format, according to the task type.
In the following example we’re using pytorch. To see an implementation of this in tensorflow, please refer to
creating VisionData guide
For pytorch, we will use our DataLoader, but we’ll create a new collate function for it, that transforms the batch to
the correct format. Then, we’ll create a deepchecks.vision.vision_data.vision_data.VisionData object, that will hold the data loader.
To learn more about the expected formats, please visit the supported tasks and formats guide.
First, we’ll create the collate function that will be used by the DataLoader. In pytorch, the collate function is used to transform the output batch to any custom format, and we’ll use that in order to transform the batch to the correct format for the checks.
from deepchecks.vision.vision_data import BatchOutputFormat
def deepchecks_collate_fn(batch) -> BatchOutputFormat:
"""Return a batch of images, labels and predictions for a batch of data. The expected format is a dictionary with
the following keys: 'images', 'labels' and 'predictions', each value is in the deepchecks format for the task.
You can also use the BatchOutputFormat class to create the output.
"""
# batch received as iterable of tuples of (image, label) and transformed to tuple of iterables of images and labels:
batch = tuple(zip(*batch))
# images:
inp = torch.stack(batch[0]).detach().numpy().transpose((0, 2, 3, 1))
mean = [0.485, 0.456, 0.406]
std = [0.229, 0.224, 0.225]
inp = std * inp + mean
images = np.clip(inp, 0, 1) * 255
#labels:
labels = batch[1]
#predictions:
logits = model.to(device)(torch.stack(batch[0]).to(device))
predictions = nn.Softmax(dim=1)(logits)
return BatchOutputFormat(images=images, labels=labels, predictions=predictions)
We have a single label here, which is the tomato class The label_map is a dictionary that maps the class id to the class name, for display purposes.
LABEL_MAP = {
0: 'ants',
1: 'bees'
}
Now that we have our updated collate function, we can recreate the dataloader in the deepchecks format, and use it to create a VisionData object:
from deepchecks.vision import VisionData
train_loader = DataLoader(train_dataset, batch_size=4, shuffle=True, collate_fn=deepchecks_collate_fn)
test_loader = DataLoader(test_dataset, batch_size=4, shuffle=True, collate_fn=deepchecks_collate_fn)
training_data = VisionData(batch_loader=train_loader, task_type='classification', label_map=LABEL_MAP)
test_data = VisionData(batch_loader=test_loader, task_type='classification', label_map=LABEL_MAP)
Making sure our data is in the correct format:#
The VisionData object automatically validates your data format and will alert you if there is a problem.
However, you can also manually view your images and labels to make sure they are in the correct format by using
the head function to conveniently visualize your data:
<deepchecks.core.serialization.html_display.HtmlDisplayableResult object at 0x7fca8e32b040>
And observe the output:
Running Deepchecks’ suite on our data and model!#
Now that we have defined the task class, we can validate the train and test data with deepchecks’ train test validation suite. This can be done with this simple few lines of code:
from deepchecks.vision.suites import train_test_validation
suite = train_test_validation()
result = suite.run(training_data, test_data, max_samples = 5000)
Processing Batches:Train:
| | 0/1 [Time: 00:00]
Processing Batches:Train:
|#####| 1/1 [Time: 00:08]
Processing Batches:Train:
|#####| 1/1 [Time: 00:08]
Computing Single Dataset Checks Train:
| | 0/0 [Time: 00:00]
Computing Single Dataset Checks Train:
| | 0/0 [Time: 00:00]
Processing Batches:Test:
| | 0/1 [Time: 00:00]
Processing Batches:Test:
|#####| 1/1 [Time: 00:05]
Processing Batches:Test:
|#####| 1/1 [Time: 00:05]
Computing Single Dataset Checks Test:
| | 0/0 [Time: 00:00]
Computing Single Dataset Checks Test:
| | 0/0 [Time: 00:00]
Computing Train Test Checks:
| | 0/6 [Time: 00:00]
Computing Train Test Checks:
| | 0/6 [Time: 00:00, Check=New Labels]
Computing Train Test Checks:
| | 0/6 [Time: 00:00, Check=Heatmap Comparison]
Computing Train Test Checks:
| | 0/6 [Time: 00:00, Check=Label Drift]
Computing Train Test Checks:
| | 0/6 [Time: 00:00, Check=Image Property Drift]
Computing Train Test Checks:
|#### | 4/6 [Time: 00:00, Check=Image Property Drift]
Computing Train Test Checks:
|#### | 4/6 [Time: 00:00, Check=Image Dataset Drift]
Computing Train Test Checks:
|#### | 4/6 [Time: 00:00, Check=Property Label Correlation Change]
Computing Train Test Checks:
|######| 6/6 [Time: 00:00, Check=Property Label Correlation Change]
Computing Train Test Checks:
|######| 6/6 [Time: 00:00, Check=Property Label Correlation Change]
We also have suites for:
data integrity
- validating a single dataset and
model evaluation -
evaluating the model’s performance.
Observing the results:#
The results can be saved as a html file with the following code:
result.save_as_html('output.html')
# Or displayed in a new window in an IDE like Pycharm:
# result.show_in_window()
'output (2).html'
Or, if working inside a notebook, the output can be displayed directly by simply printing the result object:
result
We can see that we do not have any meaningful issues with our data, and although there’s some drift between the train and test datasets (under the “Passed” section), this is not significant enough to cause any issues (and therefor is not displayed in the “Didn’t Pass” section). However, under the “Other” section, that details checks without a specific pass/fail condition, we can see that the heatmap of brightness in the images is not uniformly distributed, which means that in most images, there are brighter objects in the center of the image. This makes sense as these images of bees and ants tend to have the insects in the center of the image, but it is something to be aware of and maybe use data augmentation to fix.
Total running time of the script: ( 0 minutes 16.300 seconds)