var samples = new List<Transition>(batchSize);
for (var i = 0; i < batchSize; i++)
{
samples.Add(m_Buffer[m_CurrentIndex-1]);
samples.Add(m_Buffer[0]);
}
return samples;
var samples = m_Buffer.SampleBatch(m_Config.batchSize);
// m_ModelRunner.UpdateModel(samples);
m_ModelRunner.UpdateModel(samples);
// Update target network
// if (m_TrainingStep % m_Config.updateTargetFreq == 0)