Python numpy.resize() Examples

def img_pre_process(img):
    """
    Processes the image and returns it
    :param img: The image to be processed
    :return: Returns the processed image
    """
    ## Chop off 1/3 from the top and cut bottom 150px(which contains the head of car)
    shape = img.shape
    img = img[int(shape[0]/3):shape[0]-150, 0:shape[1]]
    ## Resize the image
    img = cv2.resize(img, (params.FLAGS.img_w, params.FLAGS.img_h), interpolation=cv2.INTER_AREA)
    ## Return the image sized as a 4D array
    return np.resize(img, (params.FLAGS.img_w, params.FLAGS.img_h, params.FLAGS.img_c))


## Process video 

def load_flo(file_path):
    """
    Read .flo file in MiddleBury format
    Code adapted from:
    http://stackoverflow.com/questions/28013200/reading-middlebury-flow-files-with-python-bytes-array-numpy

    WARNING: this will work on little-endian architectures (eg Intel x86) only!
    Args:
        file_path string: file path(absolute)
    Returns:
        flow (numpy.array): data of image in (Height, Width, 2) layout
    """

    with open(file_path, 'rb') as f:
        magic = np.fromfile(f, np.float32, count=1)
        assert(magic == 202021.25)
        w = int(np.fromfile(f, np.int32, count=1))
        h = int(np.fromfile(f, np.int32, count=1))
        # print('Reading %d x %d flo file\n' % (w, h))
        flow = np.fromfile(f, np.float32, count=2 * w * h)
        # Reshape data into 3D array (columns, rows, bands)
        # The reshape here is for visualization, the original code is (w,h,2)
        flow = np.resize(flow, (h, w, 2))

    return flow 

def read_flo_file(filename, memcached=False):
    """
    Read from Middlebury .flo file
    :param flow_file: name of the flow file
    :return: optical flow data in matrix
    """
    if memcached:
        filename = io.BytesIO(filename)
    f = open(filename, 'rb')
    magic = np.fromfile(f, np.float32, count=1)[0]
    data2d = None

    if 202021.25 != magic:
        print('Magic number incorrect. Invalid .flo file')
    else:
        w = np.fromfile(f, np.int32, count=1)[0]
        h = np.fromfile(f, np.int32, count=1)[0]
        data2d = np.fromfile(f, np.float32, count=2 * w * h)
        # reshape data into 3D array (columns, rows, channels)
        data2d = np.resize(data2d, (h, w, 2))
    f.close()
    return data2d


# fast resample layer 

def test_check_preserve_type():
    # Ensures that type float32 is preserved.
    XA = np.resize(np.arange(40), (5, 8)).astype(np.float32)
    XB = np.resize(np.arange(40), (5, 8)).astype(np.float32)

    XA_checked, XB_checked = check_pairwise_arrays(XA, None)
    assert_equal(XA_checked.dtype, np.float32)

    # both float32
    XA_checked, XB_checked = check_pairwise_arrays(XA, XB)
    assert_equal(XA_checked.dtype, np.float32)
    assert_equal(XB_checked.dtype, np.float32)

    # mismatched A
    XA_checked, XB_checked = check_pairwise_arrays(XA.astype(np.float),
                                                   XB)
    assert_equal(XA_checked.dtype, np.float)
    assert_equal(XB_checked.dtype, np.float)

    # mismatched B
    XA_checked, XB_checked = check_pairwise_arrays(XA,
                                                   XB.astype(np.float))
    assert_equal(XA_checked.dtype, np.float)
    assert_equal(XB_checked.dtype, np.float) 

def test_broadcast(size, mask, item, box):
    selection = np.resize(mask, size)

    data = np.arange(size, dtype=float)

    # Construct the expected series by taking the source
    # data or item based on the selection
    expected = Series([item if use_item else data[
        i] for i, use_item in enumerate(selection)])

    s = Series(data)
    s[selection] = box(item)
    assert_series_equal(s, expected)

    s = Series(data)
    result = s.where(~selection, box(item))
    assert_series_equal(result, expected)

    s = Series(data)
    result = s.mask(selection, box(item))
    assert_series_equal(result, expected) 

def read_flo_file(filename,verbose=False):
    """
    Read from .flo optical flow file (Middlebury format)
    :param flow_file: name of the flow file
    :return: optical flow data in matrix
    
    adapted from https://github.com/liruoteng/OpticalFlowToolkit/
    
    """
    f = open(filename, 'rb')
    magic = np.fromfile(f, np.float32, count=1)
    data2d = None

    if 202021.25 != magic:
        raise TypeError('Magic number incorrect. Invalid .flo file')
    else:
        w = np.fromfile(f, np.int32, count=1)
        h = np.fromfile(f, np.int32, count=1)
        if verbose:
            print("Reading %d x %d flow file in .flo format" % (h, w))
        data2d = np.fromfile(f, np.float32, count=int(2 * w * h))
        # reshape data into 3D array (columns, rows, channels)
        data2d = np.resize(data2d, (h[0], w[0], 2))
    f.close()
    return data2d 

def test_PLinearDropInputs_ShouldDropRightParams(self):
        dropped_index = 0

        # assume input is 2x2x2, 2 layers of 2x2
        input_shape = (2, 2, 2)
        module = pnn.PLinear(8, 10)

        old_num_features = module.in_features
        old_weight = module.weight.data.cpu().numpy()
        resized_old_weight = np.resize(old_weight, (module.out_features, *input_shape))

        module.drop_inputs(input_shape, dropped_index)
        new_shape = module.weight.size()

        # ensure that the chosen index is dropped
        expected_weight = np.resize(np.delete(resized_old_weight, dropped_index, 1), new_shape)
        output = module.weight.data.cpu().numpy()
        self.assertTrue(np.array_equal(output, expected_weight))

        # ensure num features is reduced
        self.assertTrue(module.in_features, old_num_features-1) 

def readFlow(fn):
    """ Read .flo file in Middlebury format"""
    # Code adapted from:
    # http://stackoverflow.com/questions/28013200/reading-middlebury-flow-files-with-python-bytes-array-numpy

    # WARNING: this will work on little-endian architectures (eg Intel x86) only!
    # print 'fn = %s'%(fn)
    with open(fn, 'rb') as f:
        magic = np.fromfile(f, np.float32, count=1)
        if 202021.25 != magic:
            print('Magic number incorrect. Invalid .flo file')
            return None
        else:
            w = np.fromfile(f, np.int32, count=1)
            h = np.fromfile(f, np.int32, count=1)
            # print 'Reading %d x %d flo file\n' % (w, h)
            data = np.fromfile(f, np.float32, count=2 * int(w) * int(h))
            # Reshape data into 3D array (columns, rows, bands)
            # The reshape here is for visualization, the original code is (w,h,2)
            return np.resize(data, (int(h), int(w), 2)) 

def PreprocessContentImage(path, long_edge):
    img = io.imread(path)
    logging.info("load the content image, size = %s", img.shape[:2])
    factor = float(long_edge) / max(img.shape[:2])
    new_size = (int(img.shape[0] * factor), int(img.shape[1] * factor))
    resized_img = transform.resize(img, new_size)
    sample = np.asarray(resized_img) * 256
    # swap axes to make image from (224, 224, 3) to (3, 224, 224)
    sample = np.swapaxes(sample, 0, 2)
    sample = np.swapaxes(sample, 1, 2)
    # sub mean
    sample[0, :] -= 123.68
    sample[1, :] -= 116.779
    sample[2, :] -= 103.939
    logging.info("resize the content image to %s", new_size)
    return np.resize(sample, (1, 3, sample.shape[1], sample.shape[2])) 

def locate_label(self, linecontour, labelwidth):
        """
        Find a good place to plot a label (relatively flat
        part of the contour).
        """

        nsize = len(linecontour)
        if labelwidth > 1:
            xsize = int(np.ceil(nsize / labelwidth))
        else:
            xsize = 1
        if xsize == 1:
            ysize = nsize
        else:
            ysize = int(labelwidth)

        XX = np.resize(linecontour[:, 0], (xsize, ysize))
        YY = np.resize(linecontour[:, 1], (xsize, ysize))
        #I might have fouled up the following:
        yfirst = YY[:, 0].reshape(xsize, 1)
        ylast = YY[:, -1].reshape(xsize, 1)
        xfirst = XX[:, 0].reshape(xsize, 1)
        xlast = XX[:, -1].reshape(xsize, 1)
        s = (yfirst - YY) * (xlast - xfirst) - (xfirst - XX) * (ylast - yfirst)
        L = np.sqrt((xlast - xfirst) ** 2 + (ylast - yfirst) ** 2).ravel()
        dist = np.add.reduce(([(abs(s)[i] / L[i]) for i in range(xsize)]), -1)
        x, y, ind = self.get_label_coords(dist, XX, YY, ysize, labelwidth)
        #print 'ind, x, y', ind, x, y

        # There must be a more efficient way...
        lc = [tuple(l) for l in linecontour]
        dind = lc.index((x, y))
        #print 'dind', dind
        #dind = list(linecontour).index((x,y))

        return x, y, dind 

def reconst_weights(weights, num, layer, reshape_x, reshape_y):
    weights = np.array(weights)
    weights = np.reshape(weights, (reshape_x, reshape_y))
    img = np.zeros((reshape_x, reshape_y))
    for i in range(reshape_x):
        for j in range(reshape_y):
            img[i][j] = int(interp(weights[i][j], [par.w_min, par.w_max], [0, 255]))

    img = np.resize(img, (28, 28))
    cv2.imwrite('weights/layer_' + str(layer) + '_neuron_' + str(num) + '.png', img)
    return img 

def subsample_flow(flow, subsampling):
    dtype = flow.dtype
    u, v = [pimg.fromarray(uv.squeeze()) for uv in np.split(flow / subsampling, 2, axis=-1)]
    size = tuple([int(round(wh / subsampling)) for wh in u.size])
    u, v = u.resize(size), v.resize(size)
    return np.stack([np.array(u, dtype=dtype), np.array(v, dtype=dtype)], axis=-1) 

def mesh_from_img(img):
    nv = (img.shape[0] + 1) * (img.shape[1] + 1)
    nf = img.size * 2

    v_count = 0
    f_count = 0
    V_dict = {}

    V = np.zeros([nv, 2])
    F = np.zeros([nf, 3], dtype=np.int)

    for i in range(img.shape[0]):
        for j in range(img.shape[1]):
            val = img[i, j]
            if val == 255.0:
                continue

            v_idx = []
            for v_i in [(i, j), (i + 1, j), (i, j + 1), (i + 1, j + 1)]:
                if v_i in V_dict:
                    v_idx.append(V_dict[v_i])
                else:
                    V_dict[v_i] = v_count
                    V[v_count, :] = np.array((v_i[1], -v_i[0]))
                    v_count += 1
                    v_idx.append(v_count - 1)

            v1, v2, v3, v4 = v_idx

            F[f_count, :] = np.array([v1, v2, v4])
            F[f_count + 1, :] = np.array([v1, v4, v3])
            f_count += 2

    V = np.resize(V, [v_count, 2])
    F = np.resize(F, [f_count, 3])

    return V, F 

def test_check_different_dimensions():
    # Ensure an error is raised if the dimensions are different.
    XA = np.resize(np.arange(45), (5, 9))
    XB = np.resize(np.arange(32), (4, 8))
    assert_raises(ValueError, check_pairwise_arrays, XA, XB)

    XB = np.resize(np.arange(4 * 9), (4, 9))
    assert_raises(ValueError, check_paired_arrays, XA, XB) 

def test_check_XB_returned():
    # Ensure that if XA and XB are given correctly, they return as equal.
    # Check that if XB is not None, it is returned equal.
    # Note that the second dimension of XB is the same as XA.
    XA = np.resize(np.arange(40), (5, 8))
    XB = np.resize(np.arange(32), (4, 8))
    XA_checked, XB_checked = check_pairwise_arrays(XA, XB)
    assert_array_equal(XA, XA_checked)
    assert_array_equal(XB, XB_checked)

    XB = np.resize(np.arange(40), (5, 8))
    XA_checked, XB_checked = check_paired_arrays(XA, XB)
    assert_array_equal(XA, XA_checked)
    assert_array_equal(XB, XB_checked) 

def train(self):
        #print "train step",self.time_step
        # Sample a random minibatch of N transitions from replay buffer
        minibatch = self.replay_buffer.get_batch(BATCH_SIZE)
        state_batch = np.asarray([data[0] for data in minibatch])
        program_order_batch = np.asarray([data[1] for data in minibatch])
        action_batch = np.asarray([data[2] for data in minibatch])
        reward_batch = np.asarray([data[3] for data in minibatch])
        next_state_batch = np.asarray([data[4] for data in minibatch])
        done_batch = np.asarray([data[5] for data in minibatch])
        
        # for action_dim = 1
        action_batch = np.resize(action_batch,[BATCH_SIZE,self.action_dim])

        # Calculate y_batch
        
        next_action_batch = self.actor_network.target_actions(next_state_batch,program_order_batch)
        q_value_batch = self.critic_network.target_q(next_state_batch,next_action_batch,program_order_batch)
        y_batch = []  
        for i in range(len(minibatch)): 
            if done_batch[i]:
                y_batch.append(reward_batch[i])
            else :
                y_batch.append(reward_batch[i] + GAMMA * q_value_batch[i])
        y_batch = np.resize(y_batch,[BATCH_SIZE,1])
        # Update critic by minimizing the loss L
        self.critic_network.train(y_batch,state_batch,action_batch,program_order_batch)

        # Update the actor policy using the sampled gradient:
        action_batch_for_gradients = self.actor_network.actions(state_batch,program_order_batch)
        q_gradient_batch = self.critic_network.gradients(state_batch,action_batch_for_gradients,program_order_batch)

        self.actor_network.train(q_gradient_batch,state_batch,program_order_batch)

        # Update the target networks
        self.actor_network.update_target()
        self.critic_network.update_target() 

def ringbuffer(data, length, decay=1.0, rate=44100):
    ''' Repeat data for 'length' amount of time, smoothing to reduce higher
        frequency oscillation. decay is the percent of amplitude decrease.
    '''
    phase = len(data)
    length = int(rate * length)
    out = numpy.resize(data, length)
    for i in range(phase, length):
        index = i - phase
        out[i] = (out[index] + out[index + 1]) * 0.5 * decay
    return out 

def resize_patch(patches_list, patch_shape):
    for i, patch in enumerate(patches_list):
        patches_list[i] = np.resize(patch, (1,) + (patch_shape) + (1,))
    return patches_list

# Remove all blank patches (with only background) 

def colex(a, indices, axis=1):
    """\nExtracts specified indices (a list) from passed array, along passed
    axis (column extraction is default).  BEWARE: A 1D array is presumed to be a
    column-array (and that the whole array will be returned as a column).

    Returns: the columns of a specified by indices\n"""

    if type(indices) not in [ListType,TupleType,np.ndarray]:
        indices = [indices]
    if len(np.shape(a)) == 1:
        cols = np.resize(a,[a.shape[0],1])
    else:
        cols = np.take(a,indices,axis)
    return cols 

def abut(source, *args):
    # comment: except for the repetition, this is equivalent to hstack.
    """\nLike the |Stat abut command.  It concatenates two arrays column-wise
    and returns the result.  CAUTION:  If one array is shorter, it will be
    repeated until it is as long as the other.

    Format:  abut (source, args)    where args=any # of arrays
    Returns: an array as long as the LONGEST array past, source appearing on the
    'left', arrays in <args> attached on the 'right'.\n"""

    source = asarray(source)
    if len(source.shape) == 1:
        width = 1
        source = np.resize(source,[source.shape[0],width])
    else:
        width = source.shape[1]
    for addon in args:
        if len(addon.shape) == 1:
            width = 1
            addon = np.resize(addon,[source.shape[0],width])
        else:
            width = source.shape[1]
        if len(addon) < len(source):
            addon = np.resize(addon,[source.shape[0],addon.shape[1]])
        elif len(source) < len(addon):
            source = np.resize(source,[addon.shape[0],source.shape[1]])
        source = np.concatenate((source,addon),1)
    return source 

def _reset_nest(self, data, nest=None):
        n = data[10]
        if nest is None:
            k,m = data[5],len(data[0])
            nest = m+k+1  # this is the maximum bound for nest
        else:
            if not n <= nest:
                raise ValueError("`nest` can only be increased")
        t, c, fpint, nrdata = [np.resize(data[n], nest) for n in [8,9,11,12]]

        args = data[:8] + (t,c,n,fpint,nrdata,data[13])
        data = dfitpack.fpcurf1(*args)
        return data 

def locate_label(self, linecontour, labelwidth):
        """
        Find a good place to plot a label (relatively flat
        part of the contour).
        """

        nsize = len(linecontour)
        if labelwidth > 1:
            xsize = int(np.ceil(nsize / labelwidth))
        else:
            xsize = 1
        if xsize == 1:
            ysize = nsize
        else:
            ysize = int(labelwidth)

        XX = np.resize(linecontour[:, 0], (xsize, ysize))
        YY = np.resize(linecontour[:, 1], (xsize, ysize))
        #I might have fouled up the following:
        yfirst = YY[:, 0].reshape(xsize, 1)
        ylast = YY[:, -1].reshape(xsize, 1)
        xfirst = XX[:, 0].reshape(xsize, 1)
        xlast = XX[:, -1].reshape(xsize, 1)
        s = (yfirst - YY) * (xlast - xfirst) - (xfirst - XX) * (ylast - yfirst)
        L = np.sqrt((xlast - xfirst) ** 2 + (ylast - yfirst) ** 2).ravel()
        dist = np.add.reduce(([(abs(s)[i] / L[i]) for i in range(xsize)]), -1)
        x, y, ind = self.get_label_coords(dist, XX, YY, ysize, labelwidth)
        #print 'ind, x, y', ind, x, y

        # There must be a more efficient way...
        lc = [tuple(l) for l in linecontour]
        dind = lc.index((x, y))
        #print 'dind', dind
        #dind = list(linecontour).index((x,y))

        return x, y, dind 

def __new__(cls, shape, dtype=None, buf=None, offset=0, strides=None,
                formats=None, names=None, titles=None,
                byteorder=None, aligned=False,
                mask=nomask, hard_mask=False, fill_value=None, keep_mask=True,
                copy=False,
                **options):
        #
        self = recarray.__new__(cls, shape, dtype=dtype, buf=buf, offset=offset,
                                strides=strides, formats=formats, names=names,
                                titles=titles, byteorder=byteorder,
                                aligned=aligned,)
        #
        mdtype = ma.make_mask_descr(self.dtype)
        if mask is nomask or not np.size(mask):
            if not keep_mask:
                self._mask = tuple([False] * len(mdtype))
        else:
            mask = np.array(mask, copy=copy)
            if mask.shape != self.shape:
                (nd, nm) = (self.size, mask.size)
                if nm == 1:
                    mask = np.resize(mask, self.shape)
                elif nm == nd:
                    mask = np.reshape(mask, self.shape)
                else:
                    msg = "Mask and data not compatible: data size is %i, " + \
                          "mask size is %i."
                    raise MAError(msg % (nd, nm))
                copy = True
            if not keep_mask:
                self.__setmask__(mask)
                self._sharedmask = True
            else:
                if mask.dtype == mdtype:
                    _mask = mask
                else:
                    _mask = np.array([tuple([m] * len(mdtype)) for m in mask],
                                     dtype=mdtype)
                self._mask = _mask
        return self
    #...................................................... 

def _prepare_bounds(bounds, x0):
    lb, ub = [np.asarray(b, dtype=float) for b in bounds]
    if lb.ndim == 0:
        lb = np.resize(lb, x0.shape)

    if ub.ndim == 0:
        ub = np.resize(ub, x0.shape)

    return lb, ub 

def append(self, row):
        self.n += 1
        if self.n > self._a.shape[0]:
            self._a = np.resize(self._a, (2*self._a.shape[0],)+self._a.shape[1:])
        self._a[self.n-1, ...] = row 

def _prepare_for_use(self):
        self._random_state = check_random_state(self.random_state)
        self.X, self.y = make_regression(n_samples=self.n_samples,
                                         n_features=self.n_features,
                                         n_informative=self.n_informative,
                                         n_targets=self.n_targets,
                                         random_state=self._random_state)
        self.y = np.resize(self.y, (self.y.size, self.n_targets))

        self.target_names = ["target_" + str(i) for i in range(self.n_targets)]
        self.feature_names = ["att_num_" + str(i) for i in range(self.n_num_features)]
        self.target_values = [float] * self.n_targets 

def test_zeroresize(self):
        A = np.array([[1, 2], [3, 4]])
        Ar = np.resize(A, (0,))
        assert_array_equal(Ar, np.array([]))
        assert_equal(A.dtype, Ar.dtype)

        Ar = np.resize(A, (0, 2))
        assert_equal(Ar.shape, (0, 2))

        Ar = np.resize(A, (2, 0))
        assert_equal(Ar.shape, (2, 0)) 

def test_copies(self):
        A = np.array([[1, 2], [3, 4]])
        Ar1 = np.array([[1, 2, 3, 4], [1, 2, 3, 4]])
        assert_equal(np.resize(A, (2, 4)), Ar1)

        Ar2 = np.array([[1, 2], [3, 4], [1, 2], [3, 4]])
        assert_equal(np.resize(A, (4, 2)), Ar2)

        Ar3 = np.array([[1, 2, 3], [4, 1, 2], [3, 4, 1], [2, 3, 4]])
        assert_equal(np.resize(A, (4, 3)), Ar3) 

def __new__(cls, shape, dtype=None, buf=None, offset=0, strides=None,
                formats=None, names=None, titles=None,
                byteorder=None, aligned=False,
                mask=nomask, hard_mask=False, fill_value=None, keep_mask=True,
                copy=False,
                **options):

        self = recarray.__new__(cls, shape, dtype=dtype, buf=buf, offset=offset,
                                strides=strides, formats=formats, names=names,
                                titles=titles, byteorder=byteorder,
                                aligned=aligned,)

        mdtype = ma.make_mask_descr(self.dtype)
        if mask is nomask or not np.size(mask):
            if not keep_mask:
                self._mask = tuple([False] * len(mdtype))
        else:
            mask = np.array(mask, copy=copy)
            if mask.shape != self.shape:
                (nd, nm) = (self.size, mask.size)
                if nm == 1:
                    mask = np.resize(mask, self.shape)
                elif nm == nd:
                    mask = np.reshape(mask, self.shape)
                else:
                    msg = "Mask and data not compatible: data size is %i, " + \
                          "mask size is %i."
                    raise MAError(msg % (nd, nm))
                copy = True
            if not keep_mask:
                self.__setmask__(mask)
                self._sharedmask = True
            else:
                if mask.dtype == mdtype:
                    _mask = mask
                else:
                    _mask = np.array([tuple([m] * len(mdtype)) for m in mask],
                                     dtype=mdtype)
                self._mask = _mask
        return self 

def handle(self):
        data = self.rfile.read()
        data = json.loads(data.decode('utf-8'))
        width = data['width']
        height = data['height']
        data = np.fromstring(base64.b64decode(data['image']), dtype=np.uint8)
        data = np.resize(data, (height, width, 3))
        image = Image.fromarray(data, mode='RGB')
        self.window.image = image