plotting_utils.py

# SPDX-FileCopyrightText: Copyright (c) 2022 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
# SPDX-License-Identifier: MIT
#
# Permission is hereby granted, free of charge, to any person obtaining a
# copy of this software and associated documentation files (the "Software"),
# to deal in the Software without restriction, including without limitation
# the rights to use, copy, modify, merge, publish, distribute, sublicense,
# and/or sell copies of the Software, and to permit persons to whom the
# Software is furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
# THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
# DEALINGS IN THE SOFTWARE.
import matplotlib
matplotlib.use("Agg")
import matplotlib.pylab as plt
import numpy as np


def save_figure_to_numpy(fig):
    # save it to a numpy array.
    data = np.fromstring(fig.canvas.tostring_rgb(), dtype=np.uint8, sep='')
    data = data.reshape(fig.canvas.get_width_height()[::-1] + (3,))
    return data


def plot_alignment_to_numpy(alignment, title='', info=None, phoneme_seq=None,
                            vmin=None, vmax=None):
    if phoneme_seq:
        fig, ax = plt.subplots(figsize=(15, 10))
    else:
        fig, ax = plt.subplots(figsize=(6, 4))
    im = ax.imshow(alignment, aspect='auto', origin='lower',
                   interpolation='none', vmin=vmin, vmax=vmax)
    ax.set_title(title)
    fig.colorbar(im, ax=ax)
    xlabel = 'Decoder timestep'
    if info is not None:
        xlabel += '\n\n' + info
    plt.xlabel(xlabel)
    plt.ylabel('Encoder timestep')
    plt.tight_layout()

    if phoneme_seq != None:
        # for debugging of phonemes and durs in maps. Not used by def in training code
        ax.set_yticks(np.arange(len(phoneme_seq)))
        ax.set_yticklabels(phoneme_seq)
        ax.hlines(np.arange(len(phoneme_seq)), xmin=0.0, xmax=max(ax.get_xticks()))

    fig.canvas.draw()
    data = save_figure_to_numpy(fig)
    plt.close()
    return data


def plot_gate_outputs_to_numpy(gate_targets, gate_outputs):
    fig, ax = plt.subplots(figsize=(12, 3))
    ax.scatter(range(len(gate_targets)), gate_targets, alpha=0.5,
               color='green', marker='+', s=1, label='target')
    ax.scatter(range(len(gate_outputs)), gate_outputs, alpha=0.5,
               color='red', marker='.', s=1, label='predicted')

    plt.xlabel("Frames (Green target, Red predicted)")
    plt.ylabel("Gate State")
    plt.tight_layout()

    fig.canvas.draw()
    data = save_figure_to_numpy(fig)
    plt.close()
    return data