def _on_epoch_end(
self,
epoch,
save_epochs,
save_path,
logging,
loss,
inp,
tar,
pred,
v_loss,
v_inp,
v_tar,
v_pred,
val_data,
):
self._print_info()
logging = logging.append(
{
"loss":
loss.item(),
"val_loss":
v_loss.item(),
"rel_l2_error":
l2_error(pred, tar, relative=True, squared=False)[0].item(),
"val_rel_l2_error":
l2_error(v_pred, v_tar, relative=True,
squared=False)[0].item(),
},
ignore_index=True,
sort=False,
)
print(logging.tail(1))
if (epoch + 1) % save_epochs == 0:
fig = self._create_figure(logging, loss, inp, tar, pred, v_loss,
v_inp, v_tar, v_pred)
os.makedirs(save_path, exist_ok=True)
torch.save(
self.state_dict(),
os.path.join(save_path,
"model_weights_epoch{}.pt".format(epoch + 1)),
)
logging.to_pickle(
os.path.join(save_path,
"losses_epoch{}.pkl".format(epoch + 1)), )
if fig is not None:
fig.savefig(
os.path.join(save_path,
"plot_epoch{}.png".format(epoch + 1)),
bbox_inches="tight",
)
return logging
def grid_search(x, y, rec_func, grid):
""" Grid search utility for tuning hyper-parameters. """
err_min = np.inf
grid_param = None
grid_shape = [len(val) for val in grid.values()]
err = torch.zeros(grid_shape)
for grid_val, nidx in zip(itertools.product(*grid.values()),
np.ndindex(*grid_shape)):
grid_param_cur = dict(zip(grid.keys(), grid_val))
print("Current grid parameters (" + str([cidx + 1 for cidx in nidx]) +
" / " + str(grid_shape) + "): " + str(grid_param_cur))
x_rec = rec_func(y, **grid_param_cur)
err[nidx], _ = l2_error(x_rec, x, relative=True, squared=False)
print("Rel. recovery error: {:1.2e}".format(err[nidx]), flush=True)
if err[nidx] < err_min:
grid_param = grid_param_cur
err_min = err[nidx]
return grid_param, err_min, err
def grid_search(x, y, rec_func, grid):
""" Grid search utility for tuning hyper-parameters. """
err_min = np.inf
grid_param = None
grid_shape = [len(val) for val in grid.values()]
err = torch.zeros(grid_shape)
err_psnr = torch.zeros(grid_shape)
err_ssim = torch.zeros(grid_shape)
for grid_val, nidx in zip(itertools.product(*grid.values()),
np.ndindex(*grid_shape)):
grid_param_cur = dict(zip(grid.keys(), grid_val))
print(
"Current grid parameters (" + str(list(nidx)) + " / " +
str(grid_shape) + "): " + str(grid_param_cur),
flush=True,
)
x_rec = rec_func(y, **grid_param_cur)
err[nidx], _ = l2_error(x_rec, x, relative=True, squared=False)
err_psnr[nidx] = psnr(
rotate_real(x_rec)[:, 0:1, ...],
rotate_real(x)[:, 0:1, ...],
data_range=rotate_real(x)[:, 0:1, ...].max(),
reduction="mean",
)
err_ssim[nidx] = ssim(
rotate_real(x_rec)[:, 0:1, ...],
rotate_real(x)[:, 0:1, ...],
data_range=rotate_real(x)[:, 0:1, ...].max(),
size_average=True,
)
print("Rel. recovery error: {:1.2e}".format(err[nidx]), flush=True)
print("PSNR: {:.2f}".format(err_psnr[nidx]), flush=True)
print("SSIM: {:.2f}".format(err_ssim[nidx]), flush=True)
if err[nidx] < err_min:
grid_param = grid_param_cur
err_min = err[nidx]
return grid_param, err_min, err, err_psnr, err_ssim
def err_measure_l2(x1, x2):
""" L2 error wrapper function. """
return l2_error(x1, x2, relative=True, squared=False)[1].squeeze()