luni/cvpr
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

resnet condition encoders

Browse Source
This commit is contained in:
luni
2023-02-27 17:41:47 +00:00
parent bb7128f589
commit aa3defbf07
10 changed files with 151 additions and 111 deletions
Download Patch File Download Diff File Expand all files Collapse all files
auto_encoder_static.py
+33 -29
View File
@@ -34,6 +34,7 @@ from tf_utils import check_tf_version, allow_memory_growth
from dataset import DatasetFactory, StaticDataset
from custom_utils.custom_utils import batch_to_dict, plot_3d
from models.cond_encoder import AutoEncoderNew as AutoEncoder
from custom_utils.custom_utils import Accumulator
def init_wandb():
@@ -46,9 +47,11 @@ def init_wandb():
artifact.add_file('models/cond_encoder.py')
artifact.add_file('dataset.py')
artifact.add_file('convnext/models/convnext_tf.py')
config_file_names = [key for key in sys.modules.keys() if key.startswith('configs.review.')]
config_file_names = [key for key in sys.modules.keys() if key.startswith('configs.')]
for config_file_name in config_file_names:
artifact.add_file('/'.join(config_file_name.split('.')) + '.py')
file_name = '/'.join(config_file_name.split('.')) + '.py'
if os.path.exists(file_name):
artifact.add_file(file_name)
run.log_artifact(artifact)
uv2mesh = UV2Mesh(res=256, is_skirt=config['is_skirt'], is_smpl_orig=config['is_cape'])
@@ -65,11 +68,14 @@ def train_step(batch):
e_loss = 1e-2 * tf.reduce_mean(edge_loss(mesh_pred, mesh_orig, batch['mesh_mask']))
l1_loss = tf.reduce_mean(tf.reduce_sum((tf.math.abs(batch['uv_static'] - pred)), axis=1))
ms_ssim = tf.reduce_mean(tf.image.ssim_multiscale(pred + 1, batch['uv_static'] + 1, max_val=2, power_factors=(0.5, 0.2, 0.1, 0.1, 0.1)))
# ms_ssim = tf.reduce_mean(tf.image.ssim_multiscale(pred + 1, batch['uv_static'] + 1, max_val=3, power_factors=(0.5, 0.2, 0.1, 0.1, 0.1)))
ms_ssim = tf.reduce_mean(tf.image.ssim(pred + 1, batch['uv_static'] + 1, max_val=3))
loss = 1e3 * (1 - ms_ssim)
# loss = l1_loss
metric = tf.reduce_sum(tf.math.sqrt(tf.reduce_sum(batch['mesh_mask'][:,:,tf.newaxis] * (mesh_orig - mesh_pred) ** 2, axis=-1))) / tf.reduce_sum(batch['mesh_mask'])
gradients = tape.gradient(loss, model.trainable_variables)
gradients = [(tf.clip_by_norm(grad, clip_norm=2.0)) for grad in gradients]
optimizer.apply_gradients(zip(gradients, model.trainable_variables))
return { 'train/loss': loss, 'train/edge_loss': e_loss, 'train/l1_loss': l1_loss, 'train/euclid_error': metric, 'train/ms_ssim': ms_ssim, }
@@ -84,8 +90,10 @@ def valid_step(batch):
e_loss = 1e-2 * tf.reduce_mean(edge_loss(mesh_pred, mesh_orig, batch['mesh_mask']))
l1_loss = tf.reduce_mean(tf.reduce_sum((tf.math.abs(batch['uv_static'] - pred)), axis=1))
ms_ssim = tf.reduce_mean(tf.image.ssim_multiscale(pred + 1, batch['uv_static'] + 1, max_val=2, power_factors=(0.5, 0.2, 0.1, 0.1, 0.1)))
loss = 1e3 * (1 - ms_ssim)
# ms_ssim = tf.reduce_mean(tf.image.ssim_multiscale(pred + 1, batch['uv_static'] + 1, max_val=3, power_factors=(0.5, 0.2, 0.1, 0.1, 0.1)))
ms_ssim = tf.reduce_mean(tf.image.ssim(pred + 1, batch['uv_static'] + 1, max_val=3))
loss = 1e3 * (1 - ms_ssim)
# loss = l1_loss
metric = tf.reduce_sum(tf.math.sqrt(tf.reduce_sum(batch['mesh_mask'][:,:,tf.newaxis] * (mesh_orig - mesh_pred) ** 2, axis=-1))) / tf.reduce_sum(batch['mesh_mask'])
return { 'valid/loss': loss, 'valid/edge_loss': e_loss, 'valid/l1_loss': l1_loss, 'valid/euclid_error': metric, 'valid/ms_ssim': ms_ssim, }
@@ -126,21 +134,26 @@ def train():
for batch in dataset_train:
batch = static_dataset_train.expand_batch(batch_to_dict(batch, config['signature_dynamic']))
train_loss = train_step(batch)
train_accumulator.update(train_loss)
data_valid = valid_iterator.get_next()
data_valid = static_dataset_valid.expand_batch(batch_to_dict(data_valid, config['signature_dynamic']))
valid_loss = valid_step(data_valid)
if step % 50 == 0:
data_valid = valid_iterator.get_next()
data_valid = static_dataset_valid.expand_batch(batch_to_dict(data_valid, config['signature_dynamic']))
valid_loss = valid_step(data_valid)
valid_accumulator.update(valid_loss)
to_log = dict({ **train_loss, **valid_loss, **{ 'current_step': step } })
if step % 5000 == 0 or (step < 5000 and step % 500 == 0):
train_figs = plot_everything(batch, 'train')
valid_figs = plot_everything(data_valid, 'valid')
to_log = { **to_log, **train_figs, **valid_figs }
if step == 0:
to_log = { **to_log, **{ 'model/summary': wandb.Html(get_summary_html(model)) } }
wandb.log(to_log, step=step)
if step % 1000 == 0 or (step < 1000 and step % 100 == 0):
to_log = dict({ **train_loss, **valid_loss, **{ 'current_step': step } })
if step % 5000 == 0 or (step < 5000 and step % 1000 == 0):
train_figs = plot_everything(batch, 'train')
valid_figs = plot_everything(data_valid, 'valid')
to_log = { **to_log, **train_figs, **valid_figs }
if step == 0:
to_log = { **to_log, **{ 'model/summary': wandb.Html(get_summary_html(model)) } }
wandb.log(to_log, step=step)
train_accumulator.reset()
valid_accumulator.reset()
if step % 10000 == 0:
manager_enc.save(checkpoint_number=step)
@@ -184,17 +197,6 @@ def test():
print()
class MyLRSchedule(tf.keras.optimizers.schedules.LearningRateSchedule):
def __init__(self, initial_learning_rate):
self.initial_learning_rate = initial_learning_rate
def __call__(self, step):
if step < 1000:
return self.initial_learning_rate / 10.0
else:
return self.initial_learning_rate
if __name__ == '__main__':
@@ -242,6 +244,8 @@ if __name__ == '__main__':
model : AutoEncoder = AutoEncoder(config['cond_ae'])
optimizer = tf.optimizers.Adam(learning_rate=config['lr'])
train_accumulator = Accumulator()
valid_accumulator = Accumulator()
if config['pretrained_model'] != '':
load_model(model.encoder, config['pretrained_model'] + '_enc')