Merge pull request #138 from sashafrey/master

Improvements in ThetaSnippetScore and TopTokensScore

docs/ref/messages.txt

@@ -1281,7 +1281,8 @@ Represents a configuration of a theta snippet score.
   message ThetaSnippetScoreConfig {
     optional string field_name = 1 [default = "@body"];  // obsolete in BigARTM v0.5.8
     optional string stream_name = 2 [default = "@global"];
-    repeated int32 item_id = 3 [packed = true];
+    repeated int32 item_id = 3 [packed = true];  // obsolete in BigARTM v0.5.8
+    optional int32 item_count = 4 [default = 10];
   }
 
 .. attribute:: ThetaSnippetScoreConfig.field_name
@@ -1294,8 +1295,12 @@ Represents a configuration of a theta snippet score.
 
 .. attribute:: ThetaSnippetScoreConfig.item_id
 
-   A set of values that define items for which theta snippet should be calculated.
-   Items are identified by the item id.
+   Obsolete in BigARTM v0.5.8.
+
+.. attribute:: ThetaSnippetScoreConfig.item_count
+
+   The number of items to retrieve. ThetaSnippetScore will select last *item_count* processed items
+   and return their theta vectors.
 
 
 .. _ThetaSnippetScore:

src/artm/messages.proto

@@ -315,7 +315,8 @@ message TopTokensScore {
 message ThetaSnippetScoreConfig {
   optional string field_name = 1 [default = "@body"];
   optional string stream_name = 2 [default = "@global"];
-  repeated int32 item_id = 3 [packed = true];
+  repeated int32 item_id = 3 [packed = true];  // obsolete in BigARTM v0.5.8
+  optional int32 item_count = 4 [default = 10];
 }
 
 

src/artm/score_sandbox/theta_snippet.cc

@@ -4,6 +4,7 @@
 
 #include "artm/core/exceptions.h"
 #include "artm/core/topic_model.h"
+#include "artm/core/protobuf_helpers.h"
 
 namespace artm {
 namespace score_sandbox {
@@ -18,16 +19,10 @@ void ThetaSnippet::AppendScore(
   int topics_size = topic_model.topic_size();
 
   ThetaSnippetScore theta_snippet_score;
-
-  for (int item_index = 0; item_index < config_.item_id_size(); item_index++) {
-    if (item_index == item.id()) {
-      theta_snippet_score.add_item_id(item_index);
-      auto theta_snippet_item = theta_snippet_score.add_values();
-      for (int topic_index = 0; topic_index < topics_size; ++topic_index) {
-        theta_snippet_item->add_value(theta[topic_index]);
-      }
-      break;
-    }
+  theta_snippet_score.add_item_id(item.id());
+  auto theta_snippet_item = theta_snippet_score.add_values();
+  for (int topic_index = 0; topic_index < topics_size; ++topic_index) {
+    theta_snippet_item->add_value(theta[topic_index]);
   }
 
   AppendScore(theta_snippet_score, score);
@@ -53,12 +48,28 @@ void ThetaSnippet::AppendScore(const Score& score, Score* target) {
     BOOST_THROW_EXCEPTION(::artm::core::InternalError(error_message));
   }
 
+  if (config_.item_count() <= 0 || theta_snippet_score->values_size() == 0)
+    return;
+
+  while (theta_snippet_target->values_size() < config_.item_count()) {
+    theta_snippet_target->add_item_id(-1);
+    artm::FloatArray* values_target = theta_snippet_target->add_values();
+    for (int i = 0; i < theta_snippet_score->values(0).value_size(); ++i)
+      values_target->add_value(0.0f);
+  }
+
   for (int item_index = 0; item_index < theta_snippet_score->item_id_size(); item_index++) {
-    theta_snippet_target->add_item_id(theta_snippet_score->item_id().Get(item_index));
-    auto theta_snippet_item_target = theta_snippet_target->add_values();
+    int item_id = theta_snippet_score->item_id(item_index);
+    if (item_id < 0)
+      continue;
+
+    theta_snippet_target->set_item_id(item_id % config_.item_count(), item_id);
+    artm::FloatArray* values_target = theta_snippet_target->mutable_values(item_id % config_.item_count());
+    values_target->Clear();
+
     auto theta_snippet_item_score = theta_snippet_score->values(item_index);
     for (int topic_index = 0; topic_index < theta_snippet_item_score.value_size(); topic_index++) {
-      theta_snippet_item_target->add_value(theta_snippet_item_score.value(topic_index));
+      values_target->add_value(theta_snippet_item_score.value(topic_index));
     }
   }
 }