Python caffe.NetSpec() Examples

The following are 30 code examples of caffe.NetSpec(). You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may also want to check out all available functions/classes of the module caffe , or try the search function .
Example #1
Source File: net.py    From pytorch-fcn with MIT License 6 votes vote down vote up 
def fcn(split, tops):
    n = caffe.NetSpec()
    n.color, n.hha, n.label = L.Python(module='nyud_layers',
            layer='NYUDSegDataLayer', ntop=3,
            param_str=str(dict(nyud_dir='../data/nyud', split=split,
                tops=tops, seed=1337)))
    n = modality_fcn(n, 'color', 'color')
    n = modality_fcn(n, 'hha', 'hha')
    n.score_fused = L.Eltwise(n.score_frcolor, n.score_frhha,
            operation=P.Eltwise.SUM, coeff=[0.5, 0.5])
    n.upscore = L.Deconvolution(n.score_fused,
        convolution_param=dict(num_output=40, kernel_size=64, stride=32,
            bias_term=False),
        param=[dict(lr_mult=0)])
    n.score = crop(n.upscore, n.color)
    n.loss = L.SoftmaxWithLoss(n.score, n.label,
            loss_param=dict(normalize=False, ignore_label=255))
    return n.to_proto()
Example #2
Source File: train.py    From dilation with MIT License 6 votes vote down vote up 
def make_frontend_vgg(options, is_training):
    batch_size = options.train_batch if is_training else options.test_batch
    image_path = options.train_image if is_training else options.test_image
    label_path = options.train_label if is_training else options.test_label
    net = caffe.NetSpec()
    net.data, net.label = network.make_image_label_data(
        image_path, label_path, batch_size,
        is_training, options.crop_size, options.mean)
    last = network.build_frontend_vgg(
        net, net.data, options.classes)[0]
    if options.up:
        net.upsample = network.make_upsample(last, options.classes)
        last = net.upsample
    net.loss = network.make_softmax_loss(last, net.label)
    if not is_training:
        net.accuracy = network.make_accuracy(last, net.label)
    return net.to_proto()
Example #3
Source File: train.py    From dilation with MIT License 6 votes vote down vote up 
def make_context(options, is_training):
    batch_size = options.train_batch if is_training else options.test_batch
    image_path = options.train_image if is_training else options.test_image
    label_path = options.train_label if is_training else options.test_label
    net = caffe.NetSpec()
    net.data, net.label = network.make_bin_label_data(
        image_path, label_path, batch_size,
        options.label_shape, options.label_stride)
    last = network.build_context(
        net, net.data, options.classes, options.layers)[0]
    if options.up:
        net.upsample = network.make_upsample(last, options.classes)
        last = net.upsample
    net.loss = network.make_softmax_loss(last, net.label)
    if not is_training:
        net.accuracy = network.make_accuracy(last, net.label)
    return net.to_proto()
Example #4
Source File: train.py    From deepcontext with MIT License 6 votes vote down vote up 
def gen_net(batch_size=512, use_bn=True):
    n=NetSpec();
    n.data = L.DummyData(shape={"dim":[batch_size,3,96,96]})
    n.select1 = L.DummyData(shape={"dim":[2]})
    n.select2 = L.DummyData(shape={"dim":[2]})
    n.label = L.DummyData(shape={"dim":[2]})
    caffenet_stack(n.data, n, use_bn)
    n.first = L.BatchReindex(n.relu6, n.select1)
    n.second = L.BatchReindex(n.relu6, n.select2)
    n.fc6_concat=L.Concat(n.first, n.second);

    fc_relu(n, n.fc6_concat, '7', 4096, batchnorm=use_bn);
    fc_relu(n, n.relu7, '8', 4096);
    n.fc9 = L.InnerProduct(n.relu8, num_output=8,
                            weight_filler=dict(type='xavier'));
    n.loss = L.SoftmaxWithLoss(n.fc9, n.label, loss_param=dict(normalization=P.Loss.NONE));

    prot=n.to_proto()
    prot.debug_info=True
    return prot;

# image preprocessing.  Note that the input image will modified.
Example #5
Source File: test_net_spec.py    From mix-and-match with MIT License 6 votes vote down vote up 
def lenet(batch_size):
    n = caffe.NetSpec()
    n.data, n.label = L.DummyData(shape=[dict(dim=[batch_size, 1, 28, 28]),
                                         dict(dim=[batch_size, 1, 1, 1])],
                                  transform_param=dict(scale=1./255), ntop=2)
    n.conv1 = L.Convolution(n.data, kernel_size=5, num_output=20,
        weight_filler=dict(type='xavier'))
    n.pool1 = L.Pooling(n.conv1, kernel_size=2, stride=2, pool=P.Pooling.MAX)
    n.conv2 = L.Convolution(n.pool1, kernel_size=5, num_output=50,
        weight_filler=dict(type='xavier'))
    n.pool2 = L.Pooling(n.conv2, kernel_size=2, stride=2, pool=P.Pooling.MAX)
    n.ip1 = L.InnerProduct(n.pool2, num_output=500,
        weight_filler=dict(type='xavier'))
    n.relu1 = L.ReLU(n.ip1, in_place=True)
    n.ip2 = L.InnerProduct(n.relu1, num_output=10,
        weight_filler=dict(type='xavier'))
    n.loss = L.SoftmaxWithLoss(n.ip2, n.label)
    return n.to_proto()
Example #6
Source File: net.py    From fcn with MIT License 6 votes vote down vote up 
def fcn(split, tops):
    n = caffe.NetSpec()
    n.color, n.hha, n.label = L.Python(module='nyud_layers',
            layer='NYUDSegDataLayer', ntop=3,
            param_str=str(dict(nyud_dir='../data/nyud', split=split,
                tops=tops, seed=1337)))
    n = modality_fcn(n, 'color', 'color')
    n = modality_fcn(n, 'hha', 'hha')
    n.score_fused = L.Eltwise(n.score_frcolor, n.score_frhha,
            operation=P.Eltwise.SUM, coeff=[0.5, 0.5])
    n.upscore = L.Deconvolution(n.score_fused,
        convolution_param=dict(num_output=40, kernel_size=64, stride=32,
            bias_term=False),
        param=[dict(lr_mult=0)])
    n.score = crop(n.upscore, n.color)
    n.loss = L.SoftmaxWithLoss(n.score, n.label,
            loss_param=dict(normalize=False, ignore_label=255))
    return n.to_proto()
Example #7
Source File: test_net_spec.py    From Deep-Learning-Based-Structural-Damage-Detection with MIT License 6 votes vote down vote up 
def lenet(batch_size):
    n = caffe.NetSpec()
    n.data, n.label = L.DummyData(shape=[dict(dim=[batch_size, 1, 28, 28]),
                                         dict(dim=[batch_size, 1, 1, 1])],
                                  transform_param=dict(scale=1./255), ntop=2)
    n.conv1 = L.Convolution(n.data, kernel_size=5, num_output=20,
        weight_filler=dict(type='xavier'))
    n.pool1 = L.Pooling(n.conv1, kernel_size=2, stride=2, pool=P.Pooling.MAX)
    n.conv2 = L.Convolution(n.pool1, kernel_size=5, num_output=50,
        weight_filler=dict(type='xavier'))
    n.pool2 = L.Pooling(n.conv2, kernel_size=2, stride=2, pool=P.Pooling.MAX)
    n.ip1 = L.InnerProduct(n.pool2, num_output=500,
        weight_filler=dict(type='xavier'))
    n.relu1 = L.ReLU(n.ip1, in_place=True)
    n.ip2 = L.InnerProduct(n.relu1, num_output=10,
        weight_filler=dict(type='xavier'))
    n.loss = L.SoftmaxWithLoss(n.ip2, n.label)
    return n.to_proto()
Example #8
Source File: test_coord_map.py    From Deep-Learning-Based-Structural-Damage-Detection with MIT License 6 votes vote down vote up 
def test_nd_conv(self):
        """
        ND conv maps the same way in more dimensions.
        """
        n = caffe.NetSpec()
        # define data with 3 spatial dimensions, otherwise the same net
        n.data = L.Input(shape=dict(dim=[2, 3, 100, 100, 100]))
        n.conv = L.Convolution(
            n.data, num_output=10, kernel_size=[3, 3, 3], stride=[1, 1, 1],
            pad=[0, 1, 2])
        n.pool = L.Pooling(
            n.conv, pool=P.Pooling.MAX, kernel_size=2, stride=2, pad=0)
        n.deconv = L.Deconvolution(
            n.pool, num_output=10, kernel_size=4, stride=2, pad=0)
        ax, a, b = coord_map_from_to(n.deconv, n.data)
        self.assertEquals(ax, 1)
        self.assertTrue(len(a) == len(b))
        self.assertTrue(np.all(a == 1))
        self.assertEquals(b[0] - 1, b[1])
        self.assertEquals(b[1] - 1, b[2])
Example #9
Source File: test_coord_map.py    From Deep-Learning-Based-Structural-Damage-Detection with MIT License 6 votes vote down vote up 
def coord_net_spec(ks=3, stride=1, pad=0, pool=2, dstride=2, dpad=0):
    """
    Define net spec for simple conv-pool-deconv pattern common to all
    coordinate mapping tests.
    """
    n = caffe.NetSpec()
    n.data = L.Input(shape=dict(dim=[2, 1, 100, 100]))
    n.aux = L.Input(shape=dict(dim=[2, 1, 20, 20]))
    n.conv = L.Convolution(
        n.data, num_output=10, kernel_size=ks, stride=stride, pad=pad)
    n.pool = L.Pooling(
        n.conv, pool=P.Pooling.MAX, kernel_size=pool, stride=pool, pad=0)
    # for upsampling kernel size is 2x stride
    try:
        deconv_ks = [s*2 for s in dstride]
    except:
        deconv_ks = dstride*2
    n.deconv = L.Deconvolution(
        n.pool, num_output=10, kernel_size=deconv_ks, stride=dstride, pad=dpad)
    return n
Example #10
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_prelu2(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([10, 4, 64, 64]))
        n.relu1 = L.PReLU(n.input1, channel_shared=True)
        self._test_model(*self._netspec_to_model(n, 'prelu2'))
Example #11
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_threshold2(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([6, 4, 64, 64]))
        n.abs1 = L.Threshold(n.input1, threshold=1.5)
        self._test_model(*self._netspec_to_model(n, 'threshold2'))
Example #12
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_concat(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([10, 4, 64, 64]))
        n.input2 = L.Input(shape=make_shape([10, 6, 64, 64]))
        n.input3 = L.Input(shape=make_shape([10, 8, 64, 64]))
        n.concat1 = L.Concat(n.input1, n.input2, n.input3)
        self._test_model(*self._netspec_to_model(n, 'concat'))
Example #13
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_concat2(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([10, 4, 64, 64]))
        n.input2 = L.Input(shape=make_shape([10, 6, 64, 64]))
        n.input3 = L.Input(shape=make_shape([10, 8, 64, 64]))
        n.concat1 = L.Concat(n.input1, n.input2, n.input3, axis=1)
        self._test_model(*self._netspec_to_model(n, 'concat2'))
Example #14
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_scale(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([6, 4, 64, 64]))
        n.scale1 = L.Scale(n.input1)
        self._test_model(*self._netspec_to_model(n, 'scale'))
Example #15
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_scale3(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([6, 4, 64, 64]))
        n.input2 = L.Input(shape=make_shape([6, 4, 64, 64]))
        n.scale1 = L.Scale(n.input1, n.input2, bias_term=True, axis=0)
        self._test_model(*self._netspec_to_model(n, 'scale3'))
Example #16
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_scale2(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([6, 4, 64, 64]))
        n.scale1 = L.Scale(n.input1, bias_term=True, bias_filler=make_bias_filler())
        self._test_model(*self._netspec_to_model(n, 'scale2'))
Example #17
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_scale4(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([6, 4, 64, 64]))
        n.scale1 = L.Scale(n.input1, bias_term=True, bias_filler=make_bias_filler(), axis=0)
        self._test_model(*self._netspec_to_model(n, 'scale4'))
Example #18
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_batchnorm2(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([6, 4, 64, 64]))
        n.batch_norm1 = L.BatchNorm(n.input1)
        self._test_model(*self._netspec_to_model(n, 'batchnorm2'))
Example #19
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_eltwise_sum2(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([6, 4, 64, 64]))
        n.input2 = L.Input(shape=make_shape([6, 4, 64, 64]))
        n.input3 = L.Input(shape=make_shape([6, 4, 64, 64]))
        n.sum1 = L.Eltwise(n.input1, n.input2, n.input3, coeff=[1.1, 2.2, 3.3])
        self._test_model(*self._netspec_to_model(n, 'eltwise_sum2'))

    # only testing
Example #20
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_relu2(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([10, 4, 64, 64]))
        n.relu1 = L.ReLU(n.input1, negative_slope=0.1)
        self._test_model(*self._netspec_to_model(n, 'relu2'))
Example #21
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_relu(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([10, 4, 64, 64]))
        n.relu1 = L.ReLU(n.input1)
        self._test_model(*self._netspec_to_model(n, 'relu'))
Example #22
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_flatten2(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([6, 4, 64, 64]))
        n.flatten1 = L.Flatten(n.input1, axis=-3, end_axis=-2)
        self._test_model(*self._netspec_to_model(n, 'flatten2'))
Example #23
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_threshold(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([6, 4, 64, 64]))
        n.abs1 = L.Threshold(n.input1, threshold=0.0)
        self._test_model(*self._netspec_to_model(n, 'threshold'))
Example #24
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_elu2(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([10, 4, 64, 64]))
        n.elu1 = L.ELU(n.input1, alpha=2.0)
        self._test_model(*self._netspec_to_model(n, 'elu2'))
Example #25
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_deconvolution3(self):
        CAFFE_ENGINE = 1
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([10, 4, 6, 6, 6]))
        n.deconv1 = L.Deconvolution(n.input1,
                                    convolution_param=dict(num_output=10,
                                                           kernel_size=5,
                                                           pad=1,
                                                           stride=2,
                                                           group=2,
                                                           weight_filler=make_weight_filler(),
                                                           bias_filler=make_bias_filler(),
                                                           engine=CAFFE_ENGINE))
        self._test_model(*self._netspec_to_model(n, 'deconvolution3'))
Example #26
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_deconvolution2(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([10, 4, 64, 64]))
        n.deconv1 = L.Deconvolution(n.input1,
                                    convolution_param=dict(num_output=10,
                                                           kernel_size=5,
                                                           bias_term=False,
                                                           pad_h=10,
                                                           pad_w=5,
                                                           weight_filler=make_weight_filler(),
                                                           bias_filler=make_bias_filler()))
        self._test_model(*self._netspec_to_model(n, 'deconvolution2'))
Example #27
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_deconvolution(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([10, 4, 64, 64]))
        n.deconv1 = L.Deconvolution(n.input1,
                                    convolution_param=dict(num_output=10,
                                                           kernel_size=5,
                                                           pad=10,
                                                           stride=2,
                                                           group=2,
                                                           weight_filler=make_weight_filler(),
                                                           bias_filler=make_bias_filler()))
        self._test_model(*self._netspec_to_model(n, 'deconvolution'))
Example #28
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_pooling3(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([10, 3, 64, 64]))
        n.pooling1 = L.Pooling(n.input1,
                               kernel_size=10,
                               pad_h=5,
                               pad_w=3,
                               stride_h=2,
                               stride_w=3,
                               pool=P.Pooling.MAX)
        self._test_model(*self._netspec_to_model(n, 'pooling3'))
Example #29
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_pooling2(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([10, 3, 64, 64]))
        n.pooling1 = L.Pooling(n.input1,
                               kernel_h=10,
                               kernel_w=10,
                               pad_h=5,
                               pad_w=3,
                               stride=2,
                               pool=P.Pooling.AVE)
        self._test_model(*self._netspec_to_model(n, 'pooling2'))
Example #30
Source File: caffe_layer_test_cases.py    From NNEF-Tools with Apache License 2.0 5 votes vote down vote up 
def test_group_convolution(self):
        n = caffe.NetSpec()
        n.input1 = L.Input(shape=make_shape([1, 8, 32, 32]))
        n.conv1 = L.Convolution(n.input1,
                                num_output=16,
                                kernel_size=5,
                                group=4,
                                weight_filler=make_weight_filler(),
                                bias_filler=make_bias_filler())
        self._test_model(*self._netspec_to_model(n, 'group_convolution'))