This is data of a particle-in-cell (PIC) simulation of the electrostatic two-stream instability; the PIC code used was OSIRIS [1]. This dataset was used in [2] to demonstrate the possibility of data-driven inference of the Vlasov equation (which governs the evolution of the plasma distribution function in phase space) using sparse regression techniques [2,3]. The details of the simulation are as follows: 1. The simulation is in one spatial dimension and particle velocities are restricted to that same dimension (i.e. the phase space is 1D1V) 2. The physical setup consists of two symmetrically counter-streaming populations of electrons at +/- 0.2c, where c is the speed of light; each beam has a thermal velocity of 0.04c. Each beam has equal densities, corresponding to n_0/2, where n_0 is the reference density of the simulation. There is a charge neutralizing background of immobile ions. 3. The system size is 10c/wpe, where c/wpe is the plasma skin depth associated with the reference density n_0. The simulation domain has periodic boundary conditions. 4. Each stream of electrons is represented by 10^4 particles per cell. Quadratic particle shapes are used. 5. The standard Yee solver is used to advance the electromagnetic field. Particles are advanced using the relativistic Boris algorithm. The Esirkepov scheme used for the current deposit to ensure charge conservation. No current smoothing is used. References: [1] R. A. Fonseca, L. O. Silva, F. S. Tsung, V. K. Decyk, W. Lu, C. Ren, W. B. Mori, S. Deng, S. Lee, T. Katsouleas, and J. C. Adam, Osiris: A three-dimensional, fully relativistic particle in cell code for modeling plasma based accelerators, Lect. Notes Comput. Sci. 2331, 342 (2002). [2] E. P. Alves and F. Fiuza, Data-driven discovery of reduced plasma physics models from fully kinetic simulations, Phys. Rev. Research 4, 033192 (2022).