config.toml

[master]

# GRPC port of the master node. The default value is 8086.
port = 8086

# gRPC host of the master node. The default values is "0.0.0.0".
host = "0.0.0.0"

# HTTP port of the master node. The default values is 8088.
http_port = 8088

# HTTP host of the master node. The default values is "0.0.0.0".
http_host = "0.0.0.0"

# Number of working jobs in the master node. The default value is 1.
n_jobs = 1

# Meta information timeout. The default value is 10s.
meta_timeout = "10s"

# Username for the master node dashboard.
dashboard_user_name = ""

# Password for the master node dashboard.
dashboard_password = ""

[server]

# Default number of returned items. The default value is 10.
default_n = 10

# Secret key for RESTful APIs (SSL required).
api_key = ""

# Clock error in the cluster. The default value is 5s.
clock_error = "5s"

# Insert new users while inserting feedback. The default value is true.
auto_insert_user = true

# Insert new items while inserting feedback. The default value is true.
auto_insert_item = false

[recommend]

# The cache size for recommended/popular/latest items. The default value is 10.
cache_size = 100

# Recommended cache expire time. The default value is 72h.
cache_expire = "72h"

# The context size for online recommendations. Online recommendations can't use all user feedbacks to generate
# recommendations for efficiency consideration. Instead, only the latest `context_size` feedbacks are used.
# The default value is 100.
context_size = 100

# The time-to-live (days) of active users, 0 means disabled. Recommendation won't be cached for inactive users. The default value is 0.
active_user_ttl = 0

[recommend.data_source]

# The feedback types for positive events.
positive_feedback_types = ["star","like","read>=3"]

# The feedback types for read events.
read_feedback_types = ["read"]

# The time-to-live (days) of positive feedback, 0 means disabled. The default value is 0.
positive_feedback_ttl = 0

# The time-to-live (days) of items, 0 means disabled. The default value is 0.
item_ttl = 0

[[recommend.non-personalized]]

# The name of the leaderboard.
name = "most_starred_weekly"

# The score function for items in the leaderboard.
score = "count(feedback, .FeedbackType == 'star')"

# The filter for items in the leaderboard.
filter = "(now() - item.Timestamp).Hours() < 168"

[[recommend.item-to-item]]

# The name of the item-to-item recommender.
name = "neighbors"

# The type of the item-to-item recommender. There are three types:
#   embedding: recommend by Euclidean distance of embeddings.
#   tags: recommend by number of common tags.
#   users: recommend by number of common users.
type = "embedding"

# The column of the item embeddings. Leave blank if type is "users".
column = "item.Labels.embedding"

[[recommend.user-to-user]]

# The name of the user-to-user recommender.
name = "neighbors"

# The type of the user-to-user recommender. There are three types:
#   embedding: recommend by Euclidean distance of embeddings.
#   tags: recommend by number of common tags.
#   items: recommend by number of common items.
type = "items"

[[recommend.external]]

# The name of the external recommender.
name = "trending"

# The script to fetch external recommended items. The script should return a list of item IDs.
script = """
const response = fetch("https://cdn.jsdelivr.net/gh/isboyjc/github-trending-api/data/daily/all.json");
if (!response.ok) {
  throw new Error(`${response.status} ${response.body}`);
}
const data = JSON.parse(response.body);
data["items"].map((item) => {
  return item["title"].toLowerCase().replace("/", ":");
})
"""

[recommend.collaborative]

# The type of collaborative filtering. Supported values:
#   none: disable collaborative filtering.
#   mf: matrix factorization.
type = "mf"

# The time period for model fitting. The default value is "60m".
fit_period = "60m"

# The number of epochs for model fitting. The default value is 100.
fit_epoch = 100

[recommend.collaborative.early_stopping]

# Number of epochs to wait if no improvement and then stop the training. The default value is 10.
patience = 10

[recommend.replacement]

# Replace historical items back to recommendations. The default value is false.
enable_replacement = false

# Decay the weights of replaced items from positive feedbacks. The default value is 0.8.
positive_replacement_decay = 0.8

# Decay the weights of replaced items from read feedbacks. The default value is 0.6.
read_replacement_decay = 0.6

[recommend.ranker]

# The type of the ranker. There are two types:
#   none: no ranking (default).
#   fm: factorization machines.
type = "fm"

# The time period to refresh recommendation for inactive users. The default values is 120h.
cache_expire = "120h"

# The recommenders used to fetch candidate items before ranking. The default values is all recommenders.
recommenders = ["latest", "collaborative", "non-personalized/most_starred_weekly", "item-to-item/neighbors", "user-to-user/neighbors"]

# The time period for model fitting. The default value is "60m".
fit_period = "60m"

# The number of epochs for model fitting. The default value is 100.
fit_epoch = 100

[recommend.ranker.early_stopping]

# Number of epochs to wait if no improvement and then stop the training. The default value is 10.
patience = 10

[recommend.fallback]

# The fallback recommenders are used when cached recommendation drained out. The default values is ["latest"].
recommenders = ["item-to-item/neighbors", "latest"]