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

Updated train_keras to work with SSD model directly

Browse Source 
Signed-off-by: Jim Martens <github@2martens.de>
This commit is contained in:
Jim Martens
2019-07-10 16:05:31 +02:00
parent 67c098b0d6
commit 4f6611ec6c
1 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
src/twomartens/masterthesis/ssd.py
View File

@ -507,7 +507,7 @@ def train_keras(train_generator: callable,
steps_per_epoch_train: int,
val_generator: callable,
steps_per_epoch_val: int,
ssd_model: Union[SSD, DropoutSSD],
ssd_model: tf.keras.models.Model,
weights_prefix: str,
iteration: int,
initial_epoch: int,
@ -522,7 +522,7 @@ def train_keras(train_generator: callable,
steps_per_epoch_train: number of batches per training epoch
val_generator: generator of validation data
steps_per_epoch_val: number of batches per validation epoch
ssd_model: wrapper of SSD model
ssd_model: SSD model
weights_prefix: prefix for weights directory
iteration: identifier for current training run
initial_epoch: the epoch to start training in
@ -536,7 +536,7 @@ def train_keras(train_generator: callable,
ssd_loss = keras_ssd_loss.SSDLoss()
# compile the model
ssd_model.model.compile(
ssd_model.compile(
optimizer=tf.train.AdamOptimizer(learning_rate=learning_rate_var,
beta1=0.9, beta2=0.999),
loss=ssd_loss.compute_loss,
@ -562,7 +562,7 @@ def train_keras(train_generator: callable,
if tensorboard_callback is not None:
callbacks.append(tensorboard_callback)
history = ssd_model.model.fit_generator(generator=train_generator,
history = ssd_model.fit_generator(generator=train_generator,
epochs=nr_epochs,
steps_per_epoch=steps_per_epoch_train,
validation_data=val_generator,
@ -570,8 +570,8 @@ def train_keras(train_generator: callable,
callbacks=callbacks,
initial_epoch=initial_epoch)
ssd_model.model.save(f"{checkpoint_dir}/ssd300.h5")
ssd_model.model.save_weights(f"{checkpoint_dir}/ssd300_weights.h5")
ssd_model.save(f"{checkpoint_dir}/ssd300.h5")
ssd_model.save_weights(f"{checkpoint_dir}/ssd300_weights.h5")
return history