Skip to main content

A Python wrapper for the RecomPI API

Project description

RecomPI API

A Python wrapper for the RecomPI API, providing methods to interact with the service for tracking user behavior and obtaining recommendations.

Installation

Install the RecomPI library using pip:

pip install recompi

Usage

Initialization

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.

Pushing User Behavior

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&param2=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]

Parameters

  • 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.

Getting Recommendations

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}}

Parameters

  • 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.

Explanation of Response:

The response structure ({'click': {'18': 0.25, '19': 0.75}, 'buy': {'13': 0.89, '95': 0.11}}) indicates:

  • For the 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).
  • For the 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.

Verifying API Connectivity

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]

Usage with SecureProfile

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&param2=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]

Parameters

  • 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.

Additional Details:

  • Initialization: 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.

Data Structures

Tag

Represents a tag with an ID, name, and description.

from recompi import Tag

tag = Tag(id="1", name="Technology", desc="Technology News")

Profile

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'}

SecureProfile

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'}

Location

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")

Geo

Represents geographical information with country and province.

from recompi import Geo

geo = Geo(country="Iran", province="Tehran")

Error Handling

RecomPIFieldTypeError

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)

RecomPIException

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)

Contributing

Feel free to submit issues, fork the repository, and send pull requests. Contributions are welcome!

License

This project is licensed under the MIT License.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

recompi-2.0.17.tar.gz (11.5 kB view details)

Uploaded Source

Built Distribution

recompi-2.0.17-py3-none-any.whl (9.4 kB view details)

Uploaded Python 3

File details

Details for the file recompi-2.0.17.tar.gz.

File metadata

  • Download URL: recompi-2.0.17.tar.gz
  • Upload date:
  • Size: 11.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.10.12

File hashes

Hashes for recompi-2.0.17.tar.gz
Algorithm Hash digest
SHA256 c340c0af9be76034ac755914d01ac0279099eca1f60439d05bf1645f590349e2
MD5 c7785077fac7aebf1de6d8959d4e9943
BLAKE2b-256 3e03efb3414ce07b1f3967ec7b46493066f7d76d148cec326e590aa9ddfb9c15

See more details on using hashes here.

File details

Details for the file recompi-2.0.17-py3-none-any.whl.

File metadata

  • Download URL: recompi-2.0.17-py3-none-any.whl
  • Upload date:
  • Size: 9.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.10.12

File hashes

Hashes for recompi-2.0.17-py3-none-any.whl
Algorithm Hash digest
SHA256 38b6fdf9027f0241329afdaf73e5ec9370924f962c0c552de495f87a11f0b01e
MD5 43b2a3b6dc6dd333010dd42f5e6e3aff
BLAKE2b-256 cc0bf5a1e3238434d796a46ebdd56deddb5443df3fd1878a8a64ff21e7797a63

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page