Commit 01deecd5 authored by Julian Treiber's avatar Julian Treiber

added batch_loss to CNNSupervisedTrainer template and tests

parent ed24cb65
Pipeline #235964 failed with stages
in 3 minutes and 26 seconds
......@@ -255,16 +255,17 @@ class ${tc.fileNameWithoutEnding}:
sparseLabel = loss_params['sparse_label'] if 'sparse_label' in loss_params else True
ignore_indices = [loss_params['ignore_indices']] if 'ignore_indices' in loss_params else []
loss_axis = loss_params['loss_axis'] if 'loss_axis' in loss_params else -1
batch_axis = loss_params['batch_axis'] if 'batch_axis' in loss_params else 0
if loss == 'softmax_cross_entropy':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'softmax_cross_entropy_ignore_indices':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'sigmoid_binary_cross_entropy':
loss_function = mx.gluon.loss.SigmoidBinaryCrossEntropyLoss()
elif loss == 'cross_entropy':
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel)
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'l2':
loss_function = mx.gluon.loss.L2Loss()
elif loss == 'l1':
......
......@@ -254,16 +254,17 @@ class CNNSupervisedTrainer_Alexnet:
sparseLabel = loss_params['sparse_label'] if 'sparse_label' in loss_params else True
ignore_indices = [loss_params['ignore_indices']] if 'ignore_indices' in loss_params else []
loss_axis = loss_params['loss_axis'] if 'loss_axis' in loss_params else -1
batch_axis = loss_params['batch_axis'] if 'batch_axis' in loss_params else 0
if loss == 'softmax_cross_entropy':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'softmax_cross_entropy_ignore_indices':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'sigmoid_binary_cross_entropy':
loss_function = mx.gluon.loss.SigmoidBinaryCrossEntropyLoss()
elif loss == 'cross_entropy':
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel)
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'l2':
loss_function = mx.gluon.loss.L2Loss()
elif loss == 'l1':
......
......@@ -254,16 +254,17 @@ class CNNSupervisedTrainer_CifarClassifierNetwork:
sparseLabel = loss_params['sparse_label'] if 'sparse_label' in loss_params else True
ignore_indices = [loss_params['ignore_indices']] if 'ignore_indices' in loss_params else []
loss_axis = loss_params['loss_axis'] if 'loss_axis' in loss_params else -1
batch_axis = loss_params['batch_axis'] if 'batch_axis' in loss_params else 0
if loss == 'softmax_cross_entropy':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'softmax_cross_entropy_ignore_indices':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'sigmoid_binary_cross_entropy':
loss_function = mx.gluon.loss.SigmoidBinaryCrossEntropyLoss()
elif loss == 'cross_entropy':
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel)
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'l2':
loss_function = mx.gluon.loss.L2Loss()
elif loss == 'l1':
......
......@@ -247,16 +247,17 @@ class CNNSupervisedTrainer_Invariant:
sparseLabel = loss_params['sparse_label'] if 'sparse_label' in loss_params else True
ignore_indices = [loss_params['ignore_indices']] if 'ignore_indices' in loss_params else []
loss_axis = loss_params['loss_axis'] if 'loss_axis' in loss_params else -1
batch_axis = loss_params['batch_axis'] if 'batch_axis' in loss_params else 0
if loss == 'softmax_cross_entropy':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'softmax_cross_entropy_ignore_indices':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'sigmoid_binary_cross_entropy':
loss_function = mx.gluon.loss.SigmoidBinaryCrossEntropyLoss()
elif loss == 'cross_entropy':
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel)
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'l2':
loss_function = mx.gluon.loss.L2Loss()
elif loss == 'l1':
......
......@@ -247,16 +247,17 @@ class CNNSupervisedTrainer_MultipleStreams:
sparseLabel = loss_params['sparse_label'] if 'sparse_label' in loss_params else True
ignore_indices = [loss_params['ignore_indices']] if 'ignore_indices' in loss_params else []
loss_axis = loss_params['loss_axis'] if 'loss_axis' in loss_params else -1
batch_axis = loss_params['batch_axis'] if 'batch_axis' in loss_params else 0
if loss == 'softmax_cross_entropy':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'softmax_cross_entropy_ignore_indices':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'sigmoid_binary_cross_entropy':
loss_function = mx.gluon.loss.SigmoidBinaryCrossEntropyLoss()
elif loss == 'cross_entropy':
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel)
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'l2':
loss_function = mx.gluon.loss.L2Loss()
elif loss == 'l1':
......
......@@ -247,16 +247,17 @@ class CNNSupervisedTrainer_RNNencdec:
sparseLabel = loss_params['sparse_label'] if 'sparse_label' in loss_params else True
ignore_indices = [loss_params['ignore_indices']] if 'ignore_indices' in loss_params else []
loss_axis = loss_params['loss_axis'] if 'loss_axis' in loss_params else -1
batch_axis = loss_params['batch_axis'] if 'batch_axis' in loss_params else 0
if loss == 'softmax_cross_entropy':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'softmax_cross_entropy_ignore_indices':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'sigmoid_binary_cross_entropy':
loss_function = mx.gluon.loss.SigmoidBinaryCrossEntropyLoss()
elif loss == 'cross_entropy':
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel)
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'l2':
loss_function = mx.gluon.loss.L2Loss()
elif loss == 'l1':
......
......@@ -247,16 +247,17 @@ class CNNSupervisedTrainer_RNNsearch:
sparseLabel = loss_params['sparse_label'] if 'sparse_label' in loss_params else True
ignore_indices = [loss_params['ignore_indices']] if 'ignore_indices' in loss_params else []
loss_axis = loss_params['loss_axis'] if 'loss_axis' in loss_params else -1
batch_axis = loss_params['batch_axis'] if 'batch_axis' in loss_params else 0
if loss == 'softmax_cross_entropy':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'softmax_cross_entropy_ignore_indices':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'sigmoid_binary_cross_entropy':
loss_function = mx.gluon.loss.SigmoidBinaryCrossEntropyLoss()
elif loss == 'cross_entropy':
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel)
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'l2':
loss_function = mx.gluon.loss.L2Loss()
elif loss == 'l1':
......
......@@ -247,16 +247,17 @@ class CNNSupervisedTrainer_RNNtest:
sparseLabel = loss_params['sparse_label'] if 'sparse_label' in loss_params else True
ignore_indices = [loss_params['ignore_indices']] if 'ignore_indices' in loss_params else []
loss_axis = loss_params['loss_axis'] if 'loss_axis' in loss_params else -1
batch_axis = loss_params['batch_axis'] if 'batch_axis' in loss_params else 0
if loss == 'softmax_cross_entropy':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'softmax_cross_entropy_ignore_indices':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'sigmoid_binary_cross_entropy':
loss_function = mx.gluon.loss.SigmoidBinaryCrossEntropyLoss()
elif loss == 'cross_entropy':
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel)
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'l2':
loss_function = mx.gluon.loss.L2Loss()
elif loss == 'l1':
......
......@@ -247,16 +247,17 @@ class CNNSupervisedTrainer_ResNeXt50:
sparseLabel = loss_params['sparse_label'] if 'sparse_label' in loss_params else True
ignore_indices = [loss_params['ignore_indices']] if 'ignore_indices' in loss_params else []
loss_axis = loss_params['loss_axis'] if 'loss_axis' in loss_params else -1
batch_axis = loss_params['batch_axis'] if 'batch_axis' in loss_params else 0
if loss == 'softmax_cross_entropy':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'softmax_cross_entropy_ignore_indices':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'sigmoid_binary_cross_entropy':
loss_function = mx.gluon.loss.SigmoidBinaryCrossEntropyLoss()
elif loss == 'cross_entropy':
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel)
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'l2':
loss_function = mx.gluon.loss.L2Loss()
elif loss == 'l1':
......
......@@ -247,16 +247,17 @@ class CNNSupervisedTrainer_Show_attend_tell:
sparseLabel = loss_params['sparse_label'] if 'sparse_label' in loss_params else True
ignore_indices = [loss_params['ignore_indices']] if 'ignore_indices' in loss_params else []
loss_axis = loss_params['loss_axis'] if 'loss_axis' in loss_params else -1
batch_axis = loss_params['batch_axis'] if 'batch_axis' in loss_params else 0
if loss == 'softmax_cross_entropy':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'softmax_cross_entropy_ignore_indices':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'sigmoid_binary_cross_entropy':
loss_function = mx.gluon.loss.SigmoidBinaryCrossEntropyLoss()
elif loss == 'cross_entropy':
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel)
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'l2':
loss_function = mx.gluon.loss.L2Loss()
elif loss == 'l1':
......
......@@ -247,16 +247,17 @@ class CNNSupervisedTrainer_ThreeInputCNN_M14:
sparseLabel = loss_params['sparse_label'] if 'sparse_label' in loss_params else True
ignore_indices = [loss_params['ignore_indices']] if 'ignore_indices' in loss_params else []
loss_axis = loss_params['loss_axis'] if 'loss_axis' in loss_params else -1
batch_axis = loss_params['batch_axis'] if 'batch_axis' in loss_params else 0
if loss == 'softmax_cross_entropy':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'softmax_cross_entropy_ignore_indices':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'sigmoid_binary_cross_entropy':
loss_function = mx.gluon.loss.SigmoidBinaryCrossEntropyLoss()
elif loss == 'cross_entropy':
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel)
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'l2':
loss_function = mx.gluon.loss.L2Loss()
elif loss == 'l1':
......
......@@ -254,16 +254,17 @@ class CNNSupervisedTrainer_VGG16:
sparseLabel = loss_params['sparse_label'] if 'sparse_label' in loss_params else True
ignore_indices = [loss_params['ignore_indices']] if 'ignore_indices' in loss_params else []
loss_axis = loss_params['loss_axis'] if 'loss_axis' in loss_params else -1
batch_axis = loss_params['batch_axis'] if 'batch_axis' in loss_params else 0
if loss == 'softmax_cross_entropy':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = mx.gluon.loss.SoftmaxCrossEntropyLoss(axis=loss_axis, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'softmax_cross_entropy_ignore_indices':
fromLogits = loss_params['from_logits'] if 'from_logits' in loss_params else False
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel)
loss_function = SoftmaxCrossEntropyLossIgnoreIndices(ignore_indices=ignore_indices, from_logits=fromLogits, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'sigmoid_binary_cross_entropy':
loss_function = mx.gluon.loss.SigmoidBinaryCrossEntropyLoss()
elif loss == 'cross_entropy':
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel)
loss_function = CrossEntropyLoss(axis=loss_axis, sparse_label=sparseLabel, batch_axis=batch_axis)
elif loss == 'l2':
loss_function = mx.gluon.loss.L2Loss()
elif loss == 'l1':
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment