"Scripts to train diffmahnet using 64bit, modifying Alan's original scripts"

- Finished v2 models. Trained using v1 as the initial model, and performing kdescent fitting to the 4D space {log(M(t)), t, log(M_obs), t_obs}.
    - Model names:
        cenflow_v2_0_64bit.eqx
        satflow_v2_0_64bit.eqx
    - Respective creation scripts:
        mpiexec -n 1 $watchmemory python -W ignore ~/local/diffhalos/mah/scripts/kdescent_train_64bit.py cenflow_v2_0_64bit.eqx --num-kernels 20 --num-fourier-kernels 20 --plot-loss-curve --seed 2 --include-test --initial-model cenflow_v1_064bit.eqx --steps 1200 --learning-rate 3e-5

        mpiexec -n 1 $watchmemory python -W ignore ~/local/diffhalos/mah/scripts/kdescent_train_64bit.py satflow_v2_0_64bit.eqx --num-kernels 20 --num-fourier-kernels 20 --plot-loss-curve --seed 2 --include-test --initial-model satflow_v1_0.eqx --steps 1200 --learning-rate 3e-5 --sats

- Finished v1 models, using full training data. Trained using the flowjax fitter only.
    - Model names:
        cenflow_v1_0_64bit.eqx (DONE)
        satflow_v1_0_64bit.eqx
    - Respective creation scripts:
        mpiexec -n 1 $watchmemory python -W ignore ~/local/diffhalos/mah/scripts/flowjax_train_64bit.py cenflow_v1_0_64bit.eqx --max-epochs 200 --max-patience 25 --learning-rate 8e-5 --nn-depth 8 --nn-width 48 --seed 111 --include-test --initial-model cenflow_v1_0train_64bit.eqx

        mpiexec -n 1 $watchmemory python -W ignore ~/local/diffhalos/mah/scripts/flowjax_train_64bit.py satflow_v1_0_64bit.eqx --max-epochs 200 --max-patience 25 --learning-rate 8e-5 --nn-depth 8 --nn-width 48 --seed 111 --include-test --initial-model satflow_v1_0train_64bit.eqx --sats

- Preliminary versions of v1 models, from the "train split" (~70% of the full training data) only. Trained using the flowjax fitter only.
    - Model names:
        cenflow_v1_0train_64bit.eqx (DONE)
        satflow_v1_0train_64bit.eqx (DONE)
    - Respective creation scripts:
        mpiexec -n 1 $watchmemory python -W ignore ~/local/diffhalos/mah/scripts/flowjax_train_64bit.py cenflow_v1_0train_64bit.eqx --max-epochs 400 --max-patience 30 --learning-rate 8e-5 --nn-depth 8 --nn-width 48 --seed 110

        mpiexec -n 1 $watchmemory python -W ignore ~/local/diffhalos/mah/scripts/flowjax_train_64bit.py satflow_v1_0train_64bit.eqx --max-epochs 400 --max-patience 30 --learning-rate 8e-5 --nn-depth 8 --nn-width 48 --seed 110 --sats