loss = net.loss(*res)
test_loss += (loss.cpu().detach().numpy()) * samples_in_batch
total_test_samples += samples_in_batch
status['val_loss'] = test_loss / total_test_samples
pbar.set_postfix(status)
# collect stats
stats_collector['train_loss'].append(status['train_loss'])
stats_collector['val_loss'].append(status['val_loss'])
# Produce artifacts
print('Saving artifacts to %s' % add_prefix('*'))
plots.plot_stats(stats_collector, filename=add_prefix('stats.png'))
plots.plot_embedding(*utils.get_embedding(net, original_train_loader),
filename=add_prefix('training-set-embedding.png'),
title='Training Set Embedding')
plots.plot_embedding(*utils.get_embedding(net, original_test_loader),
filename=add_prefix('testing-set-embedding.png'),
title='Test Set Embedding')
if args.animation:
imageio.mimsave(add_prefix('latent-space-development.gif'),
map(lambda x: imageio.imread(x),
sorted(glob.glob(add_prefix('gif/*.png')))),
duration=0.1)
shutil.rmtree(add_prefix('gif'))
D, X = IO.read_file(files[i]) # new data file
print('input file:', files[i])
individual = best[0]
func = toolbox.compile(expr=individual)
result = [(func(*inst[:inst_length])) for inst in D]
nresult = evals.reclass_result(X, result, prcnt)
outfile = outdir + 'model_from-' + file1 + '-using-' + nfile + '-'
outfile += str(rseed) + '-' + nfile + '-'
outfile += str(evaluate) + '-' + str(ig) + "way.txt"
print(outfile)
IO.create_file(X,nresult,outfile)
#
# plot data if selected
#
file = os.path.splitext(os.path.basename(infile))[0]
if Stats == True:
statfile = outdir + "stats-" + file + "-" + evaluate
statfile += "-" + str(rseed) + ".pdf"
print('saving stats to', statfile)
plots.plot_stats(df,statfile)
if Trees == True:
print('saving tree plot to ' + outdir + 'tree_' + str(save_seed) + '.pdf')
plots.plot_tree(best[0],save_seed,outdir)
if Fitness == True:
outfile = "fitness-" + file + "-" + evaluate + "-" + str(rseed) + ".pdf"
print('saving fitness plot to', outfile)
plots.plot_fitness(fitness,outfile)
fs_loss = f_gan.train_generator() # f stacked
fd_loss, fd_acc = f_gan.train_discriminator() # f discriminator
gs_loss = g_gan.train_generator() # g stacked
gd_loss, gd_acc = g_gan.train_discriminator() # g discriminator
leftinv_loss = cycle_gan.train_left_inverse()
rightinv_loss = cycle_gan.train_right_inverse()
if train.log:
print(f'{fs_loss:1.5f} {gs_loss:1.5f} {fd_acc:.5f} {gd_acc:.5f}')
if train.make_samples:
g_sample = f_gan.generate_sample(5)
train.append_sample(g_sample,
'samples/sample_{gen:03.0f}_{i:02.0f}.png')
train.append_stats(fs_loss, fd_loss, fd_acc, gs_loss, gd_loss, gd_acc,
leftinv_loss, rightinv_loss)
except KeyboardInterrupt:
pass
else:
train.compile_records()
plot_samples(train.samples)
plot_stats(train.stats)
save_model(f_gan.generator)
save_model(g_gan.generator)