def draw():
figs = tfmpl.create_figures(2, figsize=(4, 3), dpi=100)
figs[0].patch.set_facecolor('red')
figs[1].patch.set_facecolor((0, 1, 0))
return figs
def attention_matplotlib(self, gen_images, gen_heatmaps, gen_vectormaps):
'''
Creates a matplotlib figure and writes it to tensorboard using tf-matplotlib
gen_images: The image tensor of shape (batchsize,width,height,channels) you want to write to tensorboard
'''
r, c = 2, 5 # want to write 25 images as a 5x5 matplotlib subplot in TBD (tensorboard)
figs = tfmpl.create_figures(1, figsize=(10, 15))
cnt = 0
for idx, f in enumerate(figs):
for i in range(r):
for j in range(c):
ax = f.add_subplot(r, c, cnt + 1)
ax.set_yticklabels([])
ax.set_xticklabels([])
#import pdb;pdb.set_trace()
img_shape = gen_images[0][cnt].shape
hmps = np.squeeze(gen_heatmaps[cnt])
hmps = cv2.resize(hmps, (0, 0),
fx=8,
fy=8,
interpolation=cv2.INTER_CUBIC)
hmp = cv2.resize(hmps, (img_shape[1], img_shape[0]),
interpolation=cv2.INTER_CUBIC)
ax.imshow(gen_images[0][cnt])
ax.imshow(hmp[:, :, cnt % 5], alpha=0.5)
#ax.imshow(hmp[:,:,1],alpha=0.5)
#ax.imshow(hmp[:,:,2],alpha=0.5)
#ax.imshow(hmp[:,:,3],alpha=0.5)
#ax.imshow(hmp[:,:,4],alpha=0.5)
# writes the image at index cnt to the 5x5 grid
cnt += 1
f.tight_layout()
return figs
def draw(images):
num_figs = len(images)
height = math.ceil(num_figs / 2)
# draw figures in two column
fig = tfmpl.create_figures(1,
figsize=(12.8 if num_figs > 1 else 6.4,
4.8 * height))[0]
# pdb.set_trace()
for i in range(height):
for j in range(2):
seq = i * 2 + j + 1
if seq > num_figs:
fig.tight_layout()
return fig
if num_figs == 1:
ax = fig.add_subplot(1, 1, 1)
else:
ax = fig.add_subplot(height, 2, seq)
ax = fig.add_subplot(height, 2, seq)
ax.axis('off')
ax.imshow(images[seq - 1]['data'])
ax.set_title(images[seq - 1]['title'], fontsize=24)
fig.tight_layout()
return fig
def plotWave(audioTensor):
""" Plots audio to a wave graph """
figs = tfmpl.create_figures(BATCH_SIZE)
for i, f in enumerate(figs):
sp = f.add_subplot(111)
sp.plot(audioTensor[i, :, :])
return figs
def draw_scatter(scaled, colors):
'''Draw scatter plots. One for each color.'''
figs = tfmpl.create_figures(len(colors), figsize=(4, 4))
for idx, f in enumerate(figs):
ax = f.add_subplot(111)
ax.axis('off')
ax.scatter(scaled[:, 0], scaled[:, 1], c=colors[idx])
f.tight_layout()
return figs
def plotSpec(audioTensor):
""" Plots audio to a wave graph """
figs = tfmpl.create_figures(BATCH_SIZE)
for i, f in enumerate(figs):
sp = f.add_subplot(111)
dim = audioTensor[i, :, :]
dim = np.squeeze(dim)
sp.specgram(x=dim, NFFT=256, Fs=2)
return figs
def tf_keypoint_summary(points):
# points is BxNx2
batch_size = points.shape[0]
figs = tfmpl.create_figures(batch_size, figsize=(4,4))
try:
for idx, f in enumerate(figs):
ax = f.add_subplot(111)
ax.scatter(points[idx, :, 0], points[idx, :, 1], c='b')
f.tight_layout()
except Exception as ex:
print('-------------')
print('--- error ---')
print('-------------')
print('keypoint summary exception:', ex)
# pdb.set_trace()
return figs
def draw(images):
num_figs = len(images)
fig = tfmpl.create_figures(1, figsize=(12.8, 12.8))[0]
# pdb.set_trace()
for i in range(rows):
for j in range(cols):
seq = i * cols + j + 1
if seq > num_figs:
fig.tight_layout()
return fig
if num_figs == 1:
ax = fig.add_subplot(1, 1, 1)
else:
ax = fig.add_subplot(rows, cols, seq)
ax.axis('off')
ax.imshow(images[seq - 1, ...])
fig.tight_layout()
return fig
def draw_line(measured_rt, predicted_rt):
fig = tfmpl.create_figures(1, figsize=(16, 8))[0]
ax = fig.add_subplot(1, 2, 1)
for item in ([ax.title, ax.xaxis.label, ax.yaxis.label] + \
ax.get_xticklabels() + ax.get_yticklabels()):
item.set_fontsize(20)
# ax.axis('off')
ax.plot(measured_rt['index'], measured_rt['value'], 'b')
ax.set_title(measured_rt['title'], fontsize=24)
ax = fig.add_subplot(1, 2, 2)
for item in ([ax.title, ax.xaxis.label, ax.yaxis.label] + \
ax.get_xticklabels() + ax.get_yticklabels()):
item.set_fontsize(20)
ax.plot(predicted_rt['index'], predicted_rt['value'], 'r')
ax.set_title(predicted_rt['title'], fontsize=24)
fig.tight_layout()
return fig