Sökning: "ternary classification"
Hittade 4 uppsatser innehållade orden ternary classification.
1. Imputing connections of random gene networks from time series data using ANNs
Master-uppsats, Lunds universitet/Beräkningsbiologi och biologisk fysik - Genomgår omorganisationSammanfattning : This thesis presents the architecture of a convolutional neural network which is trained to impute the connections of randomly generated gene regulatory networks under varying amounts of regularisation. The generated gene networks are simulated from 10 different starting conditions for each set of connections in order to obtain multiple time series. LÄS MER
2. Buskars potential för urbana miljöer : utvärdering av olika buskars bladekonomi
Kandidat-uppsats, SLU/Dept. of Landscape Architecture, Planning and Management (from 130101)Sammanfattning : Artvariation i städer har i dag en låg andel buskarter, framför allt när det kommer till variation av olika buskarter och -släkten. Att öka mängden buskarter är något som i längden kan komma att vara viktig, framförallt för artvariationen (t.ex. vid sjukdomsdrabbning). LÄS MER
3. Compact ConvNets with Ternary Weights and Binary Activations
Master-uppsats, KTH/Robotik, perception och lärande, RPLSammanfattning : Compact architectures, ternary weights and binary activations are two methods suitable for making neural networks more efficient. We introduce a) a dithering binary activation which improves accuracy of ternary weight networks with binary activations by randomizing quantization error, and b) a method of implementing ternary weight networks with binary activations using binary operations. LÄS MER
4. Combining Eye- and Head-Tracking Signals for Improved Event Detection
Master-uppsats, Lunds universitet/Avdelningen för Biomedicinsk teknikSammanfattning : Analysing eye movements recorded with mobile eye-tracking devices is difficult since the eye-tracking signals are severely affected by simultaneous head and body movements. The automatic analysis methods developed for static eye-tracking systems do not take this into account and are therefore not suitable for application to data which also contains head and body movements. LÄS MER