A Python wrapper for the RecomPI API, providing methods to interact with the service for tracking user behavior and obtaining recommendations.
Install the RecomPI library using pip:
pip install recompi
Create an instance of the RecomPI class using your campaign's API key.
from recompi import RecomPI
api = RecomPI(api_key="YOUR_CAMPAIGN_TOKEN")
Campaign Token: To obtain
YOUR_CAMPAIGN_TOKEN, register on the RecomPI panel. After registration, add a campaign in the panel, and a campaign token will be generated instantly. Use this token as your API key in the code.
Use the push method to send user interaction data to the RecomPI API. You can include tags, profiles, location, and geographical information.
from recompi import RecomPI, Tag, Profile, SecureProfile, Location, Geo
api = RecomPI(api_key="YOUR_CAMPAIGN_TOKEN")
response = api.push(
label="click",
tags=[
Tag(id="1", name="Technology", desc="Technology News"),
Tag(
id="2",
name="Online Marketing Tools",
desc="Latest news on online marketing tools",
),
],
profiles=Profile("user_id", "123"),
location=Location(
ip="1.1.1.1", # Optional
url="https://www.example.com/some/path?param1=1¶m2=2",
referer="REFERER_URL", # Optional
useragent="USERAGENT", # Optional
),
geo=Geo(
country="Iran", # Optional: 'IR' or the unique country ID
province="Tehran", # Optional: Any unique string
), # Optional
)
print(response) # [success: true]
label (str): A label for the event, e.g., "click".tags (list[Tag]): List of Tag objects representing metadata.profiles (list[Profile|SecureProfile] or Profile|SecureProfile): List of Profile or SecureProfile objects or a single Profile or a single SecureProfile.location (Location, optional): A Location object containing IP, URL, referer, and user-agent information.geo (Geo, optional): A Geo object containing geographical information like country and province.Use the recom method to get recommendations based on user interactions, profiles, and geographical information.
from recompi import RecomPI, Profile, Geo
api = RecomPI(api_key="YOUR_CAMPAIGN_TOKEN")
response = api.recom(
labels=["click", "buy"],
profiles=Profile("user_id", "123"),
geo=Geo(
country="Iran", # Optional: 'IR' or unique country ID
province="Tehran", # Optional: Any unique string
),
)
print(response) # [success: true] {'click': {'18': 0.25, '19': 0.75}, 'buy': {'13': 0.89, '95': 0.11}}
labels (list[str]): List of event labels for which recommendations are requested.profiles (list[Profile|SecureProfile] or Profile|SecureProfile, optional): List of Profile or SecureProfile objects or a single Profile or a single SecureProfile.geo (Geo, optional): A Geo object containing geographical information.The response structure ({'click': {'18': 0.25, '19': 0.75}, 'buy': {'13': 0.89, '95': 0.11}}) indicates:
click label, the user is predicted to engage more with Tag.id = 19 (with a probability of 0.75) compared to Tag.id = 18 (0.25).buy label, the user is likely to make a purchase with Tag.id = 13 (0.89 probability) rather than Tag.id = 95 (0.11 probability).These predictions are derived from the user's past interactions (click and buy events) and other factors such as their profile attributes and geographical location. The recommendation engine uses this data to suggest items or actions that are likely to be of interest to the user, helping to optimize engagement or conversions based on predictive analytics.
Use the verify method to check if the API key and configuration are correct.
from recompi import RecomPI
api = RecomPI(api_key="YOUR_CAMPAIGN_TOKEN")
response = api.verify()
print("API is well configured and connected?", response.is_success()) # [success: true]
You can use SecureProfile to ensure that sensitive information such as user IDs are hashed before being sent to the API. This adds an extra layer of security by obfuscating the actual IDs.
from recompi import RecomPI, Tag, SecureProfile, Location, Geo
api = RecomPI(api_key="YOUR_CAMPAIGN_TOKEN", hash_salt="SOME_HASH_SALT")
response = api.push(
label="click",
tags=[
Tag(id="1", name="Technology", desc="Technology News"),
Tag(
id="2",
name="Online Marketing Tools",
desc="Latest news on online marketing tools",
),
],
profiles=SecureProfile(name="user_id", id="123"),
location=Location(
ip="1.1.1.1", # Optional
url="https://www.example.com/some/path?param1=1¶m2=2",
referer="REFERER_URL", # Optional
useragent="USERAGENT", # Optional
),
geo=Geo(
country="Iran", # Optional: 'IR' or the unique country ID
province="Tehran", # Optional: Any unique string
), # Optional
)
print(response) # [success: true]
label (str): A label for the event, e.g., "click".tags (list[Tag]): List of Tag objects representing metadata.profiles (list[Profile|SecureProfile] or Profile|SecureProfile): List of Profile or SecureProfile objects or a single Profile or a single SecureProfile.location (Location, optional): A Location object containing IP, URL, referer, and user-agent information.geo (Geo, optional): A Geo object containing geographical information like country and province.SecureProfile is initialized with name and id. It extends the base Profile to add the capability of hashing the id.to_json Method: This method converts the SecureProfile instance to a JSON-compatible dictionary, hashing the id if a hash_salt is provided. This is useful for securely sending profile data.recom: If you use SecureProfile in the push method, you must also use SecureProfile to retrieve data with the recom method. This ensures the consistency of hashed data between sending and retrieving.Represents a tag with an ID, name, and description.
from recompi import Tag
tag = Tag(id="1", name="Technology", desc="Technology News")
Represents a user profile with an ID and name.
from recompi import Profile
profile = Profile(name="user_id", id="123")
print(profile.to_json()) # {'user_id': '123'}
Represents a user profile with a secure ID and name.
from recompi import SecureProfile
profile = SecureProfile(name="user_id", id="123")
print(profile.to_json()) # {'user_id': 'e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855'}
print(profile.to_json("SOME_HASH_SALT")) # {'user_id': 'e6ee87b7300073f85bc86d817b6656d58443b23438faacc6737bde461ecf38cd'}
Represents a location with IP, URL, referer, and user-agent information.
from recompi import Location
location = Location(ip="1.1.1.1", url="https://www.example.com", referer="REFERER_URL", useragent="USERAGENT")
Represents geographical information with country and province.
from recompi import Geo
geo = Geo(country="Iran", province="Tehran")
Raised when an input field does not match the expected type.
from recompi import RecomPIFieldTypeError
try:
# Some operation that might raise an error
pass
except RecomPIFieldTypeError as e:
print(e)
General exception class for other RecomPI errors.
from recompi import RecomPIException
try:
# Some operation that might raise an error
pass
except RecomPIException as e:
print(e)
We welcome contributions to RecomPI! If you'd like to contribute, please follow these steps:
python3 -m venv venv
source venv/bin/activate
pip install -r requirements-dev.txt
pre-commit install --hook-type pre-commit --hook-type pre-push
For support or questions, please submit a ticket or open an issue on GitHub.
FAQs
A Python wrapper for the RecomPI API
We found that recompi demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
Research
An advanced npm supply chain attack is leveraging Ethereum smart contracts for decentralized, persistent malware control, evading traditional defenses.
Security News
Research
Attackers are impersonating Sindre Sorhus on npm with a fake 'chalk-node' package containing a malicious backdoor to compromise developers' projects.
Security News
The npm package for the LottieFiles Player web component was hit with a supply chain attack after a software engineer's npmjs credentials were compromised.