/// <summary>
/// Forward computation.
/// </summary>
/// <param name="colBottom">input blob vector.
/// </param>
/// <param name="colTop">output blob vector.
/// </param>
protected void forwardCpu(BlobCollection <T> colBottom, BlobCollection <T> colTop)
{
// Retrieve all ground truth.
m_hostGt.CopyFrom(colBottom[3]);
float[] rgfGtData = m_hostGt.GetHostDataAsFloat();
DictionaryMap <List <NormalizedBBox> > rgAllGtBboxes = m_bboxUtil.GetGroundTruth(rgfGtData, m_nNumGt, m_nBackgroundLabelId, m_bUseDifficultGt);
// Retrieve all prior bboxes. It is the same within a batch since we assume all
// images in a batch are of the same dimension.
List <List <float> > rgrgPriorVariances;
m_hostPrio.CopyFrom(colBottom[2]);
float[] rgfPriorData = m_hostPrio.GetHostDataAsFloat();
List <NormalizedBBox> rgPriorBboxes = m_bboxUtil.GetPrior(rgfPriorData, m_nNumPriors, out rgrgPriorVariances);
// Retrieve all predictions.
m_hostLoc.CopyFrom(colBottom[0]);
float[] rgfLocData = m_hostLoc.GetHostDataAsFloat();
List <LabelBBox> rgAllLocPreds = m_bboxUtil.GetLocPredictions(rgfLocData, m_nNum, m_nNumPriors, m_nLocClasses, m_bShareLocation);
// Find matches between source bboxes and ground truth bboxes.
List <DictionaryMap <List <float> > > rgAllMatchOverlaps;
m_bboxUtil.FindMatches(rgAllLocPreds, rgAllGtBboxes, rgPriorBboxes, rgrgPriorVariances, m_param.multiboxloss_param, out rgAllMatchOverlaps, out m_rgAllMatchIndices);
// Sample hard negative (and positive) examples based on mining type.
int nNumNegs;
m_nNumMatches = m_bboxUtil.MineHardExamples(colBottom[1], rgAllLocPreds, rgAllGtBboxes, rgPriorBboxes, rgrgPriorVariances, rgAllMatchOverlaps, m_param.multiboxloss_param, m_rgAllMatchIndices, m_rgrgAllNegIndices, out nNumNegs);
if (m_nNumMatches >= 1)
{
// Form data to pass on to loc_loss_layer.
List <int> rgLocShape = new List <int>()
{
1, m_nNumMatches * 4
};
m_blobLocPred.Reshape(rgLocShape);
m_blobLocGt.Reshape(rgLocShape);
m_bboxUtil.EncodeLocPrediction(rgAllLocPreds, rgAllGtBboxes, m_rgAllMatchIndices, rgPriorBboxes, rgrgPriorVariances, m_param.multiboxloss_param, m_blobLocPred, m_blobLocGt);
m_locLossLayer.Reshape(m_colLocBottom, m_colLocTop);
m_locLossLayer.Forward(m_colLocBottom, m_colLocTop);
}
else
{
m_blobLocLoss.SetData(0, 0);
}
// Form data to pass on to conf_loss_layer
if (m_bDoNegMining)
{
m_nNumConf = m_nNumMatches + nNumNegs;
}
else
{
m_nNumConf = m_nNum * m_nNumPriors;
}
if (m_nNumConf >= 1)
{
// Reshape the confidence data.
List <int> rgConfShape = new List <int>();
if (m_confLossType == MultiBoxLossParameter.ConfLossType.SOFTMAX)
{
rgConfShape.Add(m_nNumConf);
m_blobConfGt.Reshape(rgConfShape);
rgConfShape.Add(m_nNumClasses);
m_blobConfPred.Reshape(rgConfShape);
}
else if (m_confLossType == MultiBoxLossParameter.ConfLossType.LOGISTIC)
{
rgConfShape.Add(1);
rgConfShape.Add(m_nNumConf);
rgConfShape.Add(m_nNumClasses);
m_blobConfGt.Reshape(rgConfShape);
m_blobConfPred.Reshape(rgConfShape);
}
else
{
m_log.FAIL("Unknown confidence loss type.");
}
if (!m_bDoNegMining)
{
// Consider all scores.
// Share data and diff with bottom[1].
m_log.CHECK_EQ(m_blobConfPred.count(), colBottom[1].count(), "The conf pred and bottom[1] should have the same count.");
m_blobConfPred.ShareData(colBottom[1]);
}
m_blobConfGt.SetData(m_nBackgroundLabelId);
m_hostConf.CopyFrom(colBottom[1]);
float[] rgfConfData = m_hostConf.GetHostDataAsFloat();
m_bboxUtil.EncodeConfPrediction(rgfConfData, m_nNum, m_nNumPriors, m_param.multiboxloss_param, m_rgAllMatchIndices, m_rgrgAllNegIndices, rgAllGtBboxes, m_blobConfPred, m_blobConfGt);
m_confLossLayer.Reshape(m_colConfBottom, m_colConfTop);
m_confLossLayer.Forward(m_colConfBottom, m_colConfTop);
}
else
{
m_blobConfLoss.SetData(0, 0);
}
double dfNormalizer = GetNormalizer(m_param.loss_param.normalization.Value, m_nNum, m_nNumPriors, m_nNumMatches);
double dfLoss = 0;
if (m_param.propagate_down[0])
{
double dfLocLoss = Utility.ConvertVal <T>(m_blobLocLoss.GetData(0));
dfLoss += m_fLocWeight * dfLocLoss / dfNormalizer;
}
if (m_param.propagate_down[1])
{
double dfConfLoss = Utility.ConvertVal <T>(m_blobConfLoss.GetData(0));
dfLoss += dfConfLoss / dfNormalizer;
}
colTop[0].SetData(dfLoss, 0);
}
