tzrec/protos/sampler.proto

syntax = "proto2";
package tzrec.protos;

// Weighted Random Sampling ItemID not in Batch
message NegativeSampler {
    // sample data path
    // schema => id:int64 | weight:float | attrs:string
    required string input_path = 1;
    // number of negative sample
    required uint32 num_sample = 2;
    // field names of attrs in train data or eval data
    repeated string attr_fields = 3;
    // field name of item_id in train data or eval data
    required string item_id_field = 4;
    // attribute delimiter of attrs string
    optional string attr_delimiter = 5 [default=":"];
    // number of negative samples for evaluator
    optional uint32 num_eval_sample = 6 [default=0];
    // field delimiter of input data
    optional string field_delimiter = 7;
    // item id delimiter
    optional string item_id_delim = 8 [default=";"];
}


// Weighted Random Sampling ItemID not with Edge
message NegativeSamplerV2 {
    // user data path
    // schema => userid:int64 | weight:float
    required string user_input_path = 1;
    // item data path
    // schema => itemid:int64 | weight:float | attrs:string
    required string item_input_path = 2;
    // positive edge path
    // schema => userid:int64 | itemid:int64 | weight:float
    required string pos_edge_input_path = 3;
    // number of negative sample
    required uint32 num_sample = 4;
    // field names of attrs in train data or eval data
    repeated string attr_fields = 5;
    // field name of item_id in train data or eval data
    required string item_id_field = 6;
    // field name of user_id in train data or eval data
    required string user_id_field = 7;
    // attribute delimiter of attrs string
    optional string attr_delimiter = 8 [default=":"];
    // number of negative samples for evaluator
    optional uint32 num_eval_sample = 9 [default=0];
    // field delimiter of input data
    optional string field_delimiter = 10;
}

// Weighted Random Sampling ItemID not in Batch and Sampling Hard Edge
message HardNegativeSampler {
    // user data path
    // schema => userid:int64 | weight:float
    required string user_input_path = 1;
    // item data path
    // schema => itemid:int64 | weight:float | attrs:string
    required string item_input_path = 2;
    // hard negative edge path
    // schema => userid:int64 | itemid:int64 | weight:float
    required string hard_neg_edge_input_path = 3;
    // number of negative sample
    required uint32 num_sample = 4;
    // max number of hard negative sample
    required uint32 num_hard_sample = 5;
    // field names of attrs in train data or eval data
    repeated string attr_fields = 6;
    // field name of item_id in train data or eval data
    required string item_id_field = 7;
    // field name of user_id in train data or eval data
    required string user_id_field = 8;
    // attribute delimiter of attrs string
    optional string attr_delimiter = 9 [default=":"];
    // number of negative samples for evaluator
    optional uint32 num_eval_sample = 10 [default=0];
    // only works on local
    optional string field_delimiter = 11;
}

// Weighted Random Sampling ItemID not with Edge and Sampling Hard Edge
message HardNegativeSamplerV2 {
    // user data path
    // schema => userid:int64 | weight:float
    required string user_input_path = 1;
    // item data path
    // schema => itemid:int64 | weight:float | attrs:string
    required string item_input_path = 2;
    // positive edge path
    // schema => userid:int64 | itemid:int64 | weight:float
    required string pos_edge_input_path = 3;
    // hard negative edge path
    // schema => userid:int64 | itemid:int64 | weight:float
    required string hard_neg_edge_input_path = 4;
    // number of negative sample
    required uint32 num_sample = 5;
    // max number of hard negative sample
    required uint32 num_hard_sample = 6;
    // field names of attrs in train data or eval data
    repeated string attr_fields = 7;
    // field name of item_id in train data or eval data
    required string item_id_field = 8;
    // field name of user_id in train data or eval data
    required string user_id_field = 9;
    // attribute delimiter of attrs string
    optional string attr_delimiter = 10 [default=":"];
    // number of negative samples for evaluator
    optional uint32 num_eval_sample = 11 [default=0];
    // field delimiter of input data
    optional string field_delimiter = 12;
}

message TDMSampler {
    // schema => itemid:int64 | weight:float | attrs:string
    required string item_input_path = 1;
    // scheme => src_id:int64 | dst_id:int64 | weight:float
    // edge for train.
    required string edge_input_path = 2;
    // scheme => src_id:int64 | dst_id:int64 | weight:float
    // edge for retrieval beam search.
    required string predict_edge_input_path = 3;
    // field names of attrs in train data or eval data
    repeated string attr_fields = 4;
    // field name of item_id in train data or eval data
    required string item_id_field = 5;
    // the number of negative samples per layer
    repeated uint32 layer_num_sample = 6;
    // attribute delimiter of attrs string
    optional string attr_delimiter = 7 [default=":"];
    // number of negative samples for evaluator
    optional uint32 num_eval_sample = 8 [default=0];
    // field delimiter of input data
    optional string field_delimiter = 9;
    // The training process only trains a randomly selected
    // proportion of nodes in the middle layers of the tree
    optional float remain_ratio = 10 [default=1.0];
    // The type of probability for selecting and retaining
    // each layer in the middle layers of the tree
    optional string probability_type = 11 [default="UNIFORM"];
}