add numpy.uint16 support

Gs-Zhang · Jun 12, 2018 · 28a0ef4 · 28a0ef4
1 parent 4420f7a
commit 28a0ef4
Show file tree

Hide file tree

Showing 8 changed files with 41 additions and 34 deletions.
diff --git a/benchmarks/bench_motion.py b/benchmarks/bench_motion.py
@@ -21,13 +21,9 @@
 # Author: Todd Goodall <[email protected]>
 # License: BSD clause
 
-import os
 from time import time
-import numpy as np
-import skvideo.io
-import skvideo.datasets
-import skvideo.motion
 
+import skvideo.motion
 
 if __name__ == "__main__":
     # TODO: as code gets faster, make benchmark more 
@@ -51,7 +47,7 @@
     times = []
 
     for es in estimators:
-	print es
+	print(es)
         time_start = time()
 	skvideo.motion.blockMotion(vdata, method=es)
         time_end = time()

diff --git a/benchmarks/bench_vread.py b/benchmarks/bench_vread.py
@@ -18,12 +18,16 @@
 # Author: Todd Goodall <[email protected]>
 # License: BSD clause
 
-import os
 from time import time
+
 import numpy as np
-import skvideo.io
+
 import skvideo.datasets
 
+try:
+    xrange
+except NameError:
+    xrange = range
 
 if __name__ == "__main__":
 

diff --git a/doc/examples/IOcorrupt.py b/doc/examples/IOcorrupt.py
@@ -1,8 +1,7 @@
-import skvideo.io
-import skvideo.datasets
-
 import numpy as np
 
+import skvideo.datasets
+
 filename = skvideo.datasets.bigbuckbunny()
 
 vid_in = skvideo.io.FFmpegReader(filename)
@@ -19,7 +18,7 @@
       '-r': rate,
 })
 for idx, frame in enumerate(vid_in.nextFrame()):
-  print "Writing frame %d/%d" % (idx, T)
+  print("Writing frame %d/%d" % (idx, T))
   if (idx >= (T/2)) & (idx <= (T/2 + 10)):
     frame = np.random.normal(128, 128, size=frame.shape).astype(np.uint8)
   vid_out.writeFrame(frame)

diff --git a/doc/examples/motion_charts.py b/doc/examples/motion_charts.py
@@ -1,11 +1,13 @@
-import skvideo.io
-import skvideo.motion
-import skvideo.datasets
-
 import matplotlib.pyplot as plt
 import numpy as np
 import scipy.misc
 
+import skvideo.datasets
+
+try:
+    xrange
+except NameError:
+    xrange = range
 
 def getPlots(motionData):
     motionMagnitude = np.sqrt(np.sum(motionData**2, axis=2))

diff --git a/doc/examples/outputdictexample.py b/doc/examples/outputdictexample.py
@@ -1,7 +1,8 @@
-import skvideo.io
 import skvideo.measure
 import numpy as np
+import numpy as np
 
+import skvideo.measure
 
 outputfile = "test.mp4"
 outputdata = np.random.random(size=(30, 480, 640, 3)) * 255
@@ -24,4 +25,4 @@
   mSSIM += skvideo.measure.ssim(np.mean(inputdata[i], axis=2), np.mean(outputdata[i], axis=2))
 
 mSSIM /= 30.0
-print mSSIM
+print(mSSIM)
diff --git a/doc/examples/scene_parse.py b/doc/examples/scene_parse.py
@@ -1,12 +1,15 @@
-import skvideo.io
-import skvideo.datasets
+import matplotlib.pyplot as plt
+import numpy as np
+
 import skvideo.measure
 
-import numpy as np
-import matplotlib.pyplot as plt
+try:
+    xrange
+except NameError:
+    xrange = range
 
 def getPlot(edgelist1, edgelist2, t, w, h, T):
-    myDPI = 100.0
+    myDPI = 100
     fig = plt.figure(figsize=(w/myDPI, h/myDPI), dpi=myDPI)
     plt.subplot(211)
     plt.title("histogram algorithm")

diff --git a/doc/examples/sparsevid.py b/doc/examples/sparsevid.py
@@ -1,13 +1,17 @@
-import skvideo.io
-import skvideo.datasets
-
+import matplotlib.pyplot as plt
 import numpy as np
 import scipy
-import matplotlib.pyplot as plt
-from matplotlib import gridspec
 import sklearn.linear_model
+from matplotlib import gridspec
 from sklearn.feature_extraction import image
 
+import skvideo.datasets
+
+try:
+    xrange
+except NameError:
+    xrange = range
+
 np.random.seed(0)
 
 # use greedy K-SVD algorithm with OMP
@@ -25,7 +29,7 @@ def dict_step(X, C, D):
     usedidx = np.abs(C[:, k])>0
 
     if np.sum(usedidx) <= 1:
-      print "Skipping filter #%d" % (k,)
+      print("Skipping filter #%d" % (k,))
       unused_indices.append(k)
       continue
 
@@ -109,7 +113,7 @@ def plot_weights(basis):
 X = X[np.random.permutation(X.shape[0])[:10000]]
 
 for i in range(200):
-  print "Iteration %d / %d" % (i, 200)
+  print("Iteration %d / %d" % (i, 200))
   C = code_step(X, D)
   D = dict_step(X, C, D)
   frame = plot_weights(D.reshape(100, 1, 7, 7))

diff --git a/skvideo/measure/scene.py b/skvideo/measure/scene.py
@@ -4,13 +4,11 @@
 """
 
 import numpy as np
-import os
 import scipy.ndimage
 import scipy.spatial
-import time
 
-from ..utils import *
 from ..motion.gme import globalEdgeMotion
+from ..utils import *
 
 
 def _percentage_distance(canny_in, canny_out, r):
@@ -121,7 +119,7 @@ def scenedet(videodata, method='histogram', parameter1=None, min_scene_length=2)
 
         "intensity" --> Detects fast cuts using changes in colour and intensity between frames. Parameter1 is the threshold used for detection, which defaults to 30.0.
 
-    parameter1 : int
+    parameter1 : float
         Number used as a tuning parameter. See method argument for details.
 
     min_scene_length : int