Merge pull request #158 from DavidIkov/origin_master

Fix for missed errors.

Author: DavidIkov

examples/mnist-learn/models/altai/construct_network.cpp

@@ -99,8 +99,8 @@ static void create_projections(
     // Synapse creation.
     ResourceSynapseParams raster_to_input_synapse;
     raster_to_input_synapse.rule_.dopamine_plasticity_period_ = raster_to_input_synapse_dopamine_plasticity_period;
-    raster_to_input_synapse.rule_.w_min_ = raster_to_input_synapse_w_max;
-    raster_to_input_synapse.rule_.w_max_ = raster_to_input_synapse_w_min;
+    raster_to_input_synapse.rule_.w_max_ = raster_to_input_synapse_w_max;
+    raster_to_input_synapse.rule_.w_min_ = raster_to_input_synapse_w_min;
     raster_to_input_synapse.rule_.synaptic_resource_ =
         resource_from_weight(0, raster_to_input_synapse.rule_.w_min_, raster_to_input_synapse.rule_.w_max_);
     // Creating projection out of synapse.
@@ -114,21 +114,21 @@ static void create_projections(
     target_to_input_synapse_dopamine.output_type_ = knp::synapse_traits::OutputType::DOPAMINE;
     target_to_input_synapse_dopamine.weight_ = 0.179376 * 1000;
     target_to_input_synapse_dopamine.delay_ = 3;
-    auto target_to_l_proj_dopamine = constructor.add_projection(
+    auto target_to_input_proj_dopamine = constructor.add_projection(
         target_to_input_synapse_dopamine, knp::framework::projection::creators::aligned<DeltaSynapse>, pops.target_pop_,
         pops.input_pop_, false, false);
     // Marking created projection, labels will go into it.
-    network.data_.projections_from_classes_.push_back(target_to_l_proj_dopamine);
+    network.data_.projections_from_classes_.push_back(target_to_input_proj_dopamine);
 
     DeltaSynapseParams target_to_input_synapse_excitatory;
     target_to_input_synapse_excitatory.output_type_ = knp::synapse_traits::OutputType::EXCITATORY;
     target_to_input_synapse_excitatory.weight_ = -30 * 1000;
     target_to_input_synapse_excitatory.delay_ = 4;
-    auto target_to_l_proj_excitatory = constructor.add_projection(
+    auto target_to_input_proj_excitatory = constructor.add_projection(
         target_to_input_synapse_excitatory, knp::framework::projection::creators::all_to_all<DeltaSynapse>,
         pops.target_pop_, pops.input_pop_, false, false);
     // Marking created projection, labels will go into it.
-    network.data_.projections_from_classes_.push_back(target_to_l_proj_excitatory);
+    network.data_.projections_from_classes_.push_back(target_to_input_proj_excitatory);
 
     DeltaSynapseParams target_to_gate_synapse;
     target_to_gate_synapse.output_type_ = knp::synapse_traits::OutputType::EXCITATORY;
@@ -142,11 +142,11 @@ static void create_projections(
     DeltaSynapseParams input_to_output_synapse;
     input_to_output_synapse.output_type_ = knp::synapse_traits::OutputType::EXCITATORY;
     input_to_output_synapse.weight_ = 10.f * 1000;
-    knp::core::UID l_to_output_proj = constructor.add_projection(
+    knp::core::UID input_to_output_proj = constructor.add_projection(
         input_to_output_synapse, knp::framework::projection::creators::aligned<DeltaSynapse>, pops.input_pop_,
         pops.output_pop_, false, true);
     // Connecting created projection as receiver to WTA.
-    network.data_.wta_data_.back().second.push_back(l_to_output_proj);
+    network.data_.wta_data_.back().second.push_back(input_to_output_proj);
 
     DeltaSynapseParams output_to_gate_synapse;
     output_to_gate_synapse.output_type_ = knp::synapse_traits::OutputType::BLOCKING;

examples/mnist-learn/models/altai/hyperparameters.h

@@ -35,7 +35,7 @@ constexpr size_t neurons_per_column = 20;
 /// Activation threshold.
 constexpr uint16_t activation_threshold = 8531;
 /// Potential leak.
-constexpr int16_t potential_leak = static_cast<int16_t>(-(1.f - 1.f / 3.f) * 1000);
+constexpr int16_t potential_leak = 0;
 /// Negative activation threshold.
 constexpr uint16_t negative_activation_threshold = 0;
 /// Potential reset value.

knp/base-framework/impl/inference_evaluation/classification/processor.cpp

@@ -90,7 +90,8 @@ std::vector<InferenceResult> EvaluationHelper::process_inference_predictions() c
     for (size_t i = 0; i < predictions_.size(); ++i)
     {
         auto const &prediction = predictions_[i];
-        auto const &cur_data = dataset_.get_data_for_inference().first[i];
+        auto inference_begin = dataset_.get_data_for_inference().first;
+        auto const &cur_data = inference_begin[i];
 
         if (!prediction.votes_)
             ++prediction_results[cur_data.first].false_negatives_;