Python display.Image方法代碼示例

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def AddMLPModel(model, data):
    size = 28 * 28 * 1
    sizes = [size, size * 2, size * 2, 10]
    layer = data
    for i in range(len(sizes) - 1):
        layer = brew.fc(model, layer, 'dense_{}'.format(i), dim_in=sizes[i], dim_out=sizes[i + 1])
        layer = brew.relu(model, layer, 'relu_{}'.format(i))
    softmax = brew.softmax(model, layer, 'softmax')
    return softmax
    


# ### LeNet Model Definition
# 
# **Note**: This is the model used when the flag *USE_LENET_MODEL=True*
# 
# Below is another possible (and very powerful) architecture called LeNet. The primary difference from the MLP model is that LeNet is a Convolutional Neural Network (CNN), and therefore uses convolutional layers ([Conv](https://caffe2.ai/docs/operators-catalogue.html#conv)), max pooling layers ([MaxPool](https://caffe2.ai/docs/operators-catalogue.html#maxpool)), [ReLUs](https://caffe2.ai/docs/operators-catalogue.html#relu), *and* fully-connected ([FC](https://caffe2.ai/docs/operators-catalogue.html#fc)) layers. A full explanation of how a CNN works is beyond the scope of this tutorial but here are a few good resources for the curious reader:
# 
# - [Stanford cs231 CNNs for Visual Recognition](http://cs231n.github.io/convolutional-networks/) (**Recommended**)
# - [Explanation of Kernels in Image Processing](https://en.wikipedia.org/wiki/Kernel_%28image_processing%29) 
# - [Convolutional Arithmetic Tutorial](http://deeplearning.net/software/theano_versions/dev/tutorial/conv_arithmetic.html)
# 
# Notice, this function also uses Brew. However, this time we add more than just FC and Softmax layers.

# In[5]: 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def show(self, exec_widget=True):
        super(Viewer, self).show()
        self.viewAll()
        rec = self.app.desktop().screenGeometry()
        self.move(rec.width() - self.size().width(), 
                  rec.height() - self.size().height())
        if not exec_widget:
            timer = QtCore.QTimer()
            # timer.timeout.connect(self.close)
            timer.singleShot(20, self.close)
        self.app.exec_()
        try:
            from IPython.display import Image
            return Image(self.name)
        except ImportError as e:
            print(e) 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def logoNotebook(symbol, token='', version='', filter=''):
    '''This is a helper function, but the google APIs url is standardized.

    https://iexcloud.io/docs/api/#logo
    8am UTC daily

    Args:
        symbol (string); Ticker to request
        token (string); Access token
        version (string); API version
        filter (string); filters: https://iexcloud.io/docs/api/#filter-results

    Returns:
        image: result
    '''
    _raiseIfNotStr(symbol)
    url = logo(symbol, token, version, filter)['url']
    return ImageI(url=url) 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def embed_mp4_as_gif(filename):
    """ Makes a temporary gif version of an mp4 using ffmpeg for embedding in
    IPython. Intended for use in Jupyter notebooks. """
    if not os.path.exists(filename):
        print('file does not exist.')
        return

    dirname = os.path.dirname(filename)
    basename = os.path.basename(filename)
    newfile = tempfile.NamedTemporaryFile()
    newname = newfile.name + '.gif'
    if len(dirname) != 0:
        os.chdir(dirname)

    os.system('ffmpeg -i ' + basename + ' ' + newname)

    try:
        with open(newname, 'rb') as f:
            display(Image(f.read(), format='png'))
    finally:
        os.remove(newname) 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def save_to_img(src, output_path_name, src_type = "tensor", channel_order="cwd", scale = 255):
    if src_type == "tensor":
        src_arr = np.asarray(src) * scale
    elif src_type == "array":
        src_arr = src*scale
    else:
        print("save tensor error, cannot parse src type.")
        return False
    if channel_order == "cwd":
        src_arr = (np.moveaxis(src_arr,0,2)).astype(np.uint8)
    elif channel_order == "wdc":
        src_arr = src_arr.astype(np.uint8)
    else:
        print("save tensor error, cannot parse channel order.")
        return False
    src_img = PIL.Image.fromarray(src_arr)
    src_img.save(output_path_name)
    return True 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def display_graph(g, format='svg', include_asset_exists=False):
    """
    Display a TermGraph interactively from within IPython.
    """
    try:
        import IPython.display as display
    except ImportError:
        raise NoIPython("IPython is not installed.  Can't display graph.")

    if format == 'svg':
        display_cls = display.SVG
    elif format in ("jpeg", "png"):
        display_cls = partial(display.Image, format=format, embed=True)

    out = BytesIO()
    _render(g, out, format, include_asset_exists=include_asset_exists)
    return display_cls(data=out.getvalue()) 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def _jplot(*args):
    from IPython.display import Image

    with _MAGICS_LOCK:
        f, tmp = tempfile.mkstemp(".png")
        os.close(f)

        base, ext = os.path.splitext(tmp)

        img = output(
            output_formats=["png"],
            output_name_first_page_number="off",
            output_name=base,
        )

        all = [img]
        all.extend(args)

        _plot(all)

        image = Image(tmp)
        os.unlink(tmp)
        return image 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def draw(self, layout='neato', **kwargs):
        """ Draw the graph.
        Optional layout=['neato'|'dot'|'twopi'|'circo'|'fdp'|'nop']
        will use specified graphviz layout method.

        :param layout: pygraphviz layout algorithm (default: 'neato')
        :type layout: str
        """
        f, filePath = tempfile.mkstemp(suffix='.png')
        self.g.layout(prog=layout)
        self.g.draw(filePath)
        
        i = Image(filename=filePath)
        display(i)
        os.close(f)
        os.remove(filePath) 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def forward(self, input):
        # Return itself + the result of the two convolutions
        output = self.model(input) + input
        return output

# Image transformation network 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def macho_example11():
    picture = Image(filename='_static/curvas_ejemplos11.jpg')
    picture.size = (100, 100)
    return picture

# the library 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def AddLeNetModel(model, data):
    '''
    This part is the standard LeNet model: from data to the softmax prediction.
    
    For each convolutional layer we specify dim_in - number of input channels
    and dim_out - number or output channels. Also each Conv and MaxPool layer changes the
    image size. For example, kernel of size 5 reduces each side of an image by 4.

    While when we have kernel and stride sizes equal 2 in a MaxPool layer, it divides
    each side in half.
    '''
    # Image size: 28 x 28 -> 24 x 24
    conv1 = brew.conv(model, data, 'conv1', dim_in=1, dim_out=20, kernel=5)
    # Image size: 24 x 24 -> 12 x 12
    pool1 = brew.max_pool(model, conv1, 'pool1', kernel=2, stride=2)
    # Image size: 12 x 12 -> 8 x 8
    conv2 = brew.conv(model, pool1, 'conv2', dim_in=20, dim_out=50, kernel=5)
    # Image size: 8 x 8 -> 4 x 4
    pool2 = brew.max_pool(model, conv2, 'pool2', kernel=2, stride=2)
    # 50 * 4 * 4 stands for dim_out from previous layer multiplied by the image size
    # Here, the data is flattened from a tensor of dimension 50x4x4 to a vector of length 50*4*4
    fc3 = brew.fc(model, pool2, 'fc3', dim_in=50 * 4 * 4, dim_out=500)
    relu3 = brew.relu(model, fc3, 'relu3')
    # Last FC Layer
    pred = brew.fc(model, relu3, 'pred', dim_in=500, dim_out=10)
    # Softmax Layer
    softmax = brew.softmax(model, pred, 'softmax')
    
    return softmax


# The `AddModel` function below allows us to easily switch from MLP to LeNet model. Just change `USE_LENET_MODEL` at the very top of the notebook and rerun the whole thing.

# In[6]: 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def interactive( animation, size = 320 ):
	basedir = mkdtemp()
	basename = join( basedir, 'graph' )
	steps = [ Image( path ) for path in render( animation.graphs(), basename, 'png', size ) ]
	rmtree( basedir )
	slider = widgets.IntSlider( min = 0, max = len( steps ) - 1, step = 1, value = 0 )
	return widgets.interactive( lambda n: display(steps[ n ]), n = slider ) 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def display_upstream_structure(structure_dict):
    """Displays pipeline structure in the jupyter notebook.

    Args:
        structure_dict (dict): dict returned by
            :func:`~steppy.base.Step.upstream_structure`.
    """
    graph = _create_graph(structure_dict)
    plt = Image(graph.create_png())
    display(plt) 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def showarray(a, fmt='jpeg'):
    a = np.uint8(np.clip(a, 0, 255))
    f = StringIO()
    PIL.Image.fromarray(a).save(f, fmt)
    display(Image(data=f.getvalue())) 

# 需要導入模塊: from IPython import display [as 別名]
# 或者: from IPython.display import Image [as 別名]
def showarrayHQ(a, fmt='png'):
    a = np.uint8(np.clip(a, 0, 255))
    f = StringIO()
    PIL.Image.fromarray(a).save(f, fmt)
    display(Image(data=f.getvalue()))

# a couple of utility functions for converting to and from Caffe's input image layout