Bayesian evaluation of charge yields of fission fragments of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"><mml:mmultiscripts><mml:mi mathvariant="normal">U</mml:mi><mml:mprescripts /><mml:none /><mml:mn>239</mml:mn></mml:mmultiscripts></mml:math>

Recent experiments [Phys. Rev. Lett. 123, 092503 (2019); Phys. 118, 222501 (2017)] have made remarkable progress in measurements of the isotopic fission-fragment yields compound nucleus $^{239}\mathrm{U}$, which is great interests for fast-neutron reactors and benchmarks fission models. We apply Bayesian neural network (BNN) approach to learn existing evaluated charge infer incomplete $^{239}\mathrm{U}$. found that two-layer BNN improved compared single-layer overall performance. Our results support normal $^{239}\mathrm{U}$ around Sn Mo isotopes. The role odd-even effects has also been studied.