2martens/masterthesis
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Implemented mising encoding for labels

Browse Source 
Signed-off-by: Jim Martens <github@2martens.de>
This commit is contained in:
Jim Martens
2019-06-05 11:49:44 +02:00
parent 28abe6ae47
commit 292e5e3c95
1 changed files with 22 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

31
src/twomartens/masterthesis/ssd.py
View File

@ -49,6 +49,7 @@ from twomartens.masterthesis.ssd_keras.keras_loss_function import keras_ssd_loss
from twomartens.masterthesis.ssd_keras.models import keras_ssd300 from twomartens.masterthesis.ssd_keras.models import keras_ssd300
from twomartens.masterthesis.ssd_keras.models import keras_ssd300_dropout from twomartens.masterthesis.ssd_keras.models import keras_ssd300_dropout
from twomartens.masterthesis.ssd_keras.ssd_encoder_decoder import ssd_output_decoder from twomartens.masterthesis.ssd_keras.ssd_encoder_decoder import ssd_output_decoder
from twomartens.masterthesis.ssd_keras.ssd_encoder_decoder import ssd_input_encoder
K = tf.keras.backend K = tf.keras.backend
tfe = tf.contrib.eager tfe = tf.contrib.eager
@ -72,9 +73,11 @@ class SSD:
""" """
def __init__(self, mode: str, weights_path: Optional[str] = None) -> None: def __init__(self, mode: str, weights_path: Optional[str] = None) -> None:
self._model = keras_ssd300.ssd_300(image_size=IMAGE_SIZE, n_classes=N_CLASSES, self._model, self.predictor_sizes = \
mode=mode, iou_threshold=IOU_THRESHOLD, top_k=TOP_K, keras_ssd300.ssd_300(image_size=IMAGE_SIZE, n_classes=N_CLASSES,
scales=[0.07, 0.15, 0.33, 0.51, 0.69, 0.87, 1.05]) mode=mode, iou_threshold=IOU_THRESHOLD, top_k=TOP_K,
scales=[0.07, 0.15, 0.33, 0.51, 0.69, 0.87, 1.05],
return_predictor_sizes=True)
self.mode = mode self.mode = mode
# load existing weights # load existing weights
@ -107,10 +110,14 @@ class DropoutSSD:
""" """
def __init__(self, mode: str, weights_path: Optional[str] = None) -> None: def __init__(self, mode: str, weights_path: Optional[str] = None) -> None:
self._model = keras_ssd300_dropout.ssd_300_dropout(image_size=IMAGE_SIZE, n_classes=N_CLASSES, self._model, self.predictor_sizes = \
dropout_rate=DROPOUT_RATE, mode=mode, keras_ssd300_dropout.ssd_300_dropout(image_size=IMAGE_SIZE,
iou_threshold=IOU_THRESHOLD, top_k=TOP_K, n_classes=N_CLASSES,
scales=[0.07, 0.15, 0.33, 0.51, 0.69, 0.87, 1.05]) dropout_rate=DROPOUT_RATE, mode=mode,
iou_threshold=IOU_THRESHOLD,
top_k=TOP_K,
scales=[0.07, 0.15, 0.33, 0.51, 0.69, 0.87, 1.05],
return_predictor_sizes=True)
self.mode = mode self.mode = mode
# load existing weights # load existing weights
@ -391,6 +398,10 @@ def train(dataset: tf.data.Dataset,
latest_checkpoint = tf.train.latest_checkpoint(checkpoint_dir) latest_checkpoint = tf.train.latest_checkpoint(checkpoint_dir)
checkpoint = tf.train.Checkpoint(**checkpointables) checkpoint = tf.train.Checkpoint(**checkpointables)
checkpoint.restore(latest_checkpoint) checkpoint.restore(latest_checkpoint)
# input encoder
input_encoder = ssd_input_encoder.SSDInputEncoder(IMAGE_SIZE[0], IMAGE_SIZE[1],
N_CLASSES, checkpointables['ssd'].predictor_sizes)
def _get_last_epoch(epoch_var: tf.Variable, **kwargs) -> int: def _get_last_epoch(epoch_var: tf.Variable, **kwargs) -> int:
return int(epoch_var) return int(epoch_var)
@ -406,7 +417,7 @@ def train(dataset: tf.data.Dataset,
for epoch in range(nr_epochs - previous_epochs): for epoch in range(nr_epochs - previous_epochs):
_epoch = epoch + previous_epochs _epoch = epoch + previous_epochs
outputs = _train_one_epoch(_epoch, dataset, **checkpointables) outputs = _train_one_epoch(_epoch, dataset, input_encoder, **checkpointables)
if verbose: if verbose:
print(( print((
@ -427,6 +438,7 @@ def train(dataset: tf.data.Dataset,
def _train_one_epoch(epoch: int, def _train_one_epoch(epoch: int,
dataset: tf.data.Dataset, dataset: tf.data.Dataset,
input_encoder: ssd_input_encoder.SSDInputEncoder,
ssd: tf.keras.Model, ssd: tf.keras.Model,
ssd_optimizer: tf.train.Optimizer, ssd_optimizer: tf.train.Optimizer,
global_step: tf.Variable, global_step: tf.Variable,
@ -441,10 +453,11 @@ def _train_one_epoch(epoch: int,
# go through data set # go through data set
for x, y in dataset: for x, y in dataset:
encoded_ground_truth = input_encoder(y)
ssd_train_loss = _train_ssd_step(ssd=ssd, ssd_train_loss = _train_ssd_step(ssd=ssd,
optimizer=ssd_optimizer, optimizer=ssd_optimizer,
inputs=x, inputs=x,
ground_truth=y, ground_truth=encoded_ground_truth,
global_step=global_step) global_step=global_step)
ssd_loss_avg(ssd_train_loss) ssd_loss_avg(ssd_train_loss)
global_step.assign_add(1) global_step.assign_add(1)