This is a simple library that helps in getting some data from coursera
which can be used by content aggregators
pip install ce-scraper
this feature gives us the first 10 results of the search
to perform a simple search..
first import the search_courses
module and call the
from ce_scraper import search_courses
courses=search_courses.simple_search(search_phrase='machine learning',type='json')
print(courses)
##Output
[
{
"course_title": "Machine Learning",
"partner": "Stanford University",
"rating_value": "4.9",
"rating_count": "154,848",
"enrollment_numbers": "3.9m",
"course_difficulty": "Mixed",
"type": "COURSE"
},
{
"course_title": "Deep Learning",
"partner": "DeepLearning.AI",
"rating_value": "4.8",
"rating_count": "117,652",
"enrollment_numbers": "990k",
"course_difficulty": "Intermediate",
"type": "SPECIALIZATION"
},
{
"course_title": "Machine Learning",
"partner": "University of Washington",
"rating_value": "4.6",
"rating_count": "14,496",
"enrollment_numbers": "400k",
"course_difficulty": "Intermediate",
"type": "SPECIALIZATION"
},
{
"course_title": "Mathematics for Machine Learning",
"partner": "Imperial College London",
"rating_value": "4.6",
"rating_count": "10,371",
"enrollment_numbers": "250k",
"course_difficulty": "Beginner",
"type": "SPECIALIZATION"
},
{
"course_title": "IBM Machine Learning",
"partner": "IBM",
"rating_value": "4.6",
"rating_count": "211",
"enrollment_numbers": "11k",
"course_difficulty": "Intermediate",
"type": "PROFESSIONAL CERTIFICATE"
},
{
"course_title": "Applied Data Science with Python",
"partner": "University of Michigan",
"rating_value": "4.5",
"rating_count": "28,245",
"enrollment_numbers": "700k",
"course_difficulty": "Intermediate",
"type": "SPECIALIZATION"
},
{
"course_title": "Machine Learning for All",
"partner": "University of London",
"rating_value": "4.7",
"rating_count": "2,290",
"enrollment_numbers": "80k",
"course_difficulty": "Beginner",
"type": "COURSE"
},
{
"course_title": "IBM Data Science",
"partner": "IBM",
"rating_value": "4.6",
"rating_count": "68,872",
"enrollment_numbers": "740k",
"course_difficulty": "Beginner",
"type": "PROFESSIONAL CERTIFICATE"
},
{
"course_title": "Naive Bayes 101: Resume Selection with Machine Learning",
"partner": "Coursera Project Network",
"rating_value": "Not Available",
"rating_count": "Not Available",
"enrollment_numbers": "Not Available",
"course_difficulty": "Intermediate",
"type": "GUIDED PROJECT"
},
{
"course_title": "Advanced Machine Learning",
"partner": "HSE University",
"rating_value": "4.4",
"rating_count": "3,701",
"enrollment_numbers": "300k",
"course_difficulty": "Advanced",
"type": "SPECIALIZATION"
}
]
search_phrase
is the key word that is used to search the courses
type
is the type of file we want from the search result.
these are the supported values of type
on this release.
i. list
ii. json
iii. dataframe
To perform the advanced search we call the advanced_search
function in the search course module
with a dictionary of filters as an argument.
this feature currently supports only page number as a filter.
from ce_scraper import search_courses
courses=search_courses.advanced_search(search_phrase='deep learning', filters={'pages': 2}, type=list)
print(courses)
The above code searches with key word deep learning
and returns the courses
from two pages result on the coursera page which is around 20 courses.
More filters and features will be included in the next release of this library.