Home » Django » Difference between Django's annotate and aggregate methods?

Difference between Django's annotate and aggregate methods?

Posted by: admin November 30, 2017 Leave a comment


Django’s QuerySet has two methods, annotate and aggregate. The documentation says that:

Unlike aggregate(), annotate() is not a terminal clause. The output of the annotate() clause is a QuerySet.

Is there any other difference between them? If not, then why does aggregate exist?


I would focus on the example queries rather than your quote from the documentation. Aggregate calculates values for the entire queryset. Annotate calculates summary values for each item in the queryset.


>>> Book.objects.aggregate(average_price=Avg('price'))
{'average_price': 34.35}

Returns a dictionary containing the average price of all books in the queryset.


>>> q = Book.objects.annotate(num_authors=Count('authors'))
>>> q[0].num_authors
>>> q[1].num_authors

q is the queryset of books, but each book has been annotated with the number of authors.


That’s the main difference, but aggregates also work on a grander scale than annotations. Annotations are inherently related to individual items in a queryset. If you run an Count annotation on a something like a many-to-many field, you’ll get a separate count for each member of the queryset (as an added attribute). If you were to do the same with an aggregation, however, it would attempt to count every relationship on every member of the queryset, even duplicates, and return that as just one value.


Aggregate generate result (summary) values over an entire QuerySet.
Aggregate operate over the rowset to get a single value from the rowset.(For example sum of all prices in the rowset). Aggregate is applied on entire QuerySet and it generate result (summary) values over an entire QuerySet.

In Model:

`class Books(models.Model):
    name = models.CharField(max_length=100)
    pages = models.IntegerField()
    price = models.DecimalField(max_digits=5, decimal_places=3)

In shell:

>>> Books.objects.all().aggregate(Avg('price'))
# Above code will give the Average of the price Column 
>>> {'price__avg': 34.35}


Annotate generate an independent summary for each object in a QuerySet.(We can say it iterate each object in a QuerySet and apply operation)

In Model:

class Video(models.Model):
    name = models.CharField(max_length=52, verbose_name='Name')
    video = models.FileField(upload_to=document_path, verbose_name='Upload 
    created_by = models.ForeignKey(User, verbose_name='Created by', 
    user_likes = models.ManyToManyField(UserProfile, null=True, 
                  blank=True, help_text='User can like once', 
                         verbose_name='Like by')

In view

videos = Video.objects.values(‘id’, ‘name’,’video’).annotate(Count(‘user_likes’,distinct=True)

**In view it will count the likes for each video, **