Python layers.Attention() Examples

def _setup_layers(self):
        """
        Creating layers of model.
        1. GCN layers.
        2. Primary capsules.
        3. Attention
        4. Graph capsules.
        5. Class capsules.
        6. Reconstruction layers.
        """
        self._setup_base_layers()
        self._setup_primary_capsules()
        self._setup_attention()
        self._setup_graph_capsules()
        self._setup_class_capsule()
        self._setup_reconstruction_layers() 

def _setup_attention(self):
        """
        Creating attention layer.
        """
        self.attention = Attention(self.args.gcn_layers*self.args.capsule_dimensions,
                                   self.args.inner_attention_dimension) 

def __init__(self, name='ra', nimg=2048, na=512, nh=512, nw=512, nout=8843, npatch=30, model_file=None):
        self.name = name
        if model_file is not None:
            with h5py.File(model_file, 'r') as f:
                nimg = f.attrs['nimg']
                na = f.attrs['na']
                nh = f.attrs['nh']
                nw = f.attrs['nw']
                nout = f.attrs['nout']
                # npatch = f.attrs['npatch']
        self.config = {'nimg': nimg, 'na': na, 'nh': nh, 'nw': nw, 'nout': nout, 'npatch': npatch}

        # word embedding layer
        self.embedding = Embedding(n_emb=nout, dim_emb=nw, name=self.name+'@embedding')

        # initialization mlp layer
        self.init_mlp = MLP(layer_sizes=[na, 2*nh], output_type='tanh', name=self.name+'@init_mlp')
        self.proj_mlp = MLP(layer_sizes=[nimg, na], output_type='tanh', name=self.name+'@proj_mlp')

        # lstm
        self.lstm = BasicLSTM(dim_x=na+nw, dim_h=nh, name=self.name+'@lstm')

        # prediction mlp
        self.pred_mlp = MLP(layer_sizes=[na+nh+nw, nout], output_type='softmax', name=self.name+'@pred_mlp')

        # attention layer
        self.attention = Attention(dim_item=na, dim_context=na+nw+nh, hsize=nh, name=self.name+'@attention')

        # inputs
        cap = T.imatrix('cap')
        img = T.tensor3('img')
        self.inputs = [cap, img]

        # go through sequence
        feat = self.proj_mlp.compute(img)
        init_e = feat.mean(axis=1)
        init_state = T.concatenate([init_e, self.init_mlp.compute(init_e)], axis=-1)
        (state, self.p, loss, self.alpha), _ = theano.scan(fn=self.scan_func,
                                                           sequences=[cap[0:-1, :], cap[1:, :]],
                                                           outputs_info=[init_state, None, None, None],
                                                           non_sequences=[feat])

        # loss function
        loss = T.mean(loss)
        self.costs = [loss]

        # layers and parameters
        self.layers = [self.embedding, self.init_mlp, self.proj_mlp, self.attention, self.lstm, self.pred_mlp]
        self.params = sum([l.params for l in self.layers], [])

        # load weights from file, if model_file is not None
        if model_file is not None:
            self.load_weights(model_file)

        # these functions and variables are used in test stage
        self._init_func = None
        self._step_func = None
        self._proj_func = None
        self._feat_shared = theano.shared(np.zeros((1, npatch, na)).astype(theano.config.floatX)) 

def __init__(self, name='rass', nimg=2048, nh=512, nw=512, na=512, nout=8843, ns=80, npatch=30, model_file=None):
        self.name = name
        if model_file is not None:
            with h5py.File(model_file, 'r') as f:
                nimg = f.attrs['nimg']
                nh = f.attrs['nh']
                nw = f.attrs['nw']
                na = f.attrs['na']
                ns = f.attrs['ns']
                nout = f.attrs['nout']
        self.config = {'nimg': nimg, 'nh': nh, 'nw': nw, 'na': na, 'nout': nout, 'ns': ns, 'npatch': npatch}

        # word embedding layer
        self.embedding = Embedding(n_emb=nout, dim_emb=nw, name=self.name+'@embedding')

        # initialization mlp layer
        self.init_mlp = MLP(layer_sizes=[na, 2*nh], output_type='tanh', name=self.name+'@init_mlp')
        self.proj_mlp = MLP(layer_sizes=[nimg, na], output_type='tanh', name=self.name+'@proj_mlp')

        # attention layer
        self.attention = Attention(dim_item=na, dim_context=na+nw+nh, hsize=nh, name=self.name+'@attention')

        # lstm
        self.lstm = BasicLSTM(dim_x=na+nw+ns, dim_h=nh, name=self.name+'@lstm')

        # prediction mlp
        self.pred_mlp = MLP(layer_sizes=[na+nh+nw+ns, nout], output_type='softmax', name=self.name+'@pred_mlp')

        # inputs
        cap = T.imatrix('cap')
        img = T.tensor3('img')
        scene = T.matrix('scene')
        self.inputs = [cap, img, scene]

        # go through sequence
        feat = self.proj_mlp.compute(img)
        init_e = feat.mean(axis=1)
        init_state = T.concatenate([init_e, self.init_mlp.compute(init_e)], axis=-1)
        (state, self.p, loss, self.alpha), _ = theano.scan(fn=self.scan_func,
                                                           sequences=[cap[0:-1, :], cap[1:, :]],
                                                           outputs_info=[init_state, None, None, None],
                                                           non_sequences=[feat, scene])

        # loss function
        loss = T.mean(loss)
        self.costs = [loss]

        # layers and parameters
        self.layers = [self.embedding, self.init_mlp, self.proj_mlp, self.attention, self.lstm, self.pred_mlp]
        self.params = sum([l.params for l in self.layers], [])

        # load weights from file, if model_file is not None
        if model_file is not None:
            self.load_weights(model_file)

        # initialization for test stage
        self._init_func = None
        self._step_func = None
        self._proj_func = None
        self._feat_shared = theano.shared(np.zeros((1, npatch, na)).astype(theano.config.floatX))
        self._scene_shared = theano.shared(np.zeros((1, ns)).astype(theano.config.floatX))