Replace opencv-python with more standard Pillow

examples/manual_reporting.py

@@ -48,11 +48,13 @@ logger.report_scatter2d("example_scatter", "series_xy", iteration=1, scatter=sca
 scatter3d = np.random.randint(10, size=(10, 3))
 logger.report_scatter3d("example_scatter_3d", "series_xyz", iteration=1, scatter=scatter3d)
 
-# report image
+# report images
-m = np.eye(256, 256, dtype=np.uint8)*255
-logger.report_image_and_upload("test case", "image uint", iteration=1, matrix=m)
 m = np.eye(256, 256, dtype=np.float)
 logger.report_image_and_upload("test case", "image float", iteration=1, matrix=m)
+m = np.eye(256, 256, dtype=np.uint8)*255
+logger.report_image_and_upload("test case", "image uint8", iteration=1, matrix=m)
+m = np.concatenate((np.atleast_3d(m), np.zeros((256, 256, 2), dtype=np.uint8)), axis=2)
+logger.report_image_and_upload("test case", "image color red", iteration=1, matrix=m)
 
 # flush reports (otherwise it will be flushed in the background, every couple of seconds)
 logger.flush()

examples/tensorboard_toy.py

@@ -2,7 +2,7 @@
 #
 import tensorflow as tf
 import numpy as np
-import cv2
+from PIL import Image
 
 from trains import Task
 task = Task.init(project_name='examples', task_name='tensorboard toy example')
@@ -46,7 +46,8 @@ all_combined = tf.concat(all_distributions, 0)
 tf.summary.histogram("all_combined", all_combined)
 
 # convert to 4d [batch, col, row, RGB-channels]
-image = cv2.imread('./samples/picasso.jpg')[:, :, ::-1]
+image_open = Image.open('./samples/picasso.jpg')
+image = np.asarray(image_open)
 image_gray = image[:, :, 0][np.newaxis, :, :, np.newaxis]
 image_rgba = np.concatenate((image, 255*np.atleast_3d(np.ones(shape=image.shape[:2], dtype=np.uint8))), axis=2)
 image_rgba = image_rgba[np.newaxis, :, :, :]

requirements.txt

@@ -13,8 +13,8 @@ humanfriendly>=2.1
 jsonmodels>=2.2
 jsonschema>=2.6.0
 numpy>=1.10
-opencv-python>=3.2.0.8
 pathlib2>=2.3.0
+Pillow>=4.1.1
 pigar>=0.9.2
 plotly>=3.9.0
 psutil>=3.4.2

trains/backend_interface/metrics/events.py

@@ -3,13 +3,13 @@ import time
 from threading import Lock
 
 import attr
-import cv2
 import numpy as np
 import pathlib2
 import six
-from ...backend_api.services import events
+from PIL import Image
 from six.moves.urllib.parse import urlparse, urlunparse
 
+from ...backend_api.services import events
 from ...config import config
 
 
@@ -248,12 +248,10 @@ class UploadEvent(MetricsEventAdapter):
                 image_data = np.reshape(image_data, (height, width))
 
             # serialize image
-            _, img_bytes = cv2.imencode(
+            image = Image.fromarray(image_data)
-                self._format, image_data,
+            output = six.BytesIO()
-                params=(cv2.IMWRITE_JPEG_QUALITY, self._quality),
+            image_format = Image.registered_extensions().get(self._format.lower(), 'JPEG')
-            )
+            image.save(output, format=image_format, quality=self._quality)
-
-            output = six.BytesIO(img_bytes.tostring())
             output.seek(0)
         else:
             local_file = self._local_image_path

trains/backend_interface/metrics/reporter.py

@@ -203,7 +203,7 @@ class Reporter(InterfaceBase, AbstractContextManager, SetupUploadMixin, AsyncMan
         :type value: int
         :param path: A path to an image file. Required unless matrix is provided.
         :type path: str
-        :param matrix: A 3D numpy.ndarray object containing image data (BGR). Required unless filename is provided.
+        :param matrix: A 3D numpy.ndarray object containing image data (RGB). Required unless filename is provided.
         :type matrix: str
         :param max_image_history: maximum number of image to store per metric/variant combination
         use negative value for unlimited. default is set in global configuration (default=5)
@@ -488,7 +488,7 @@ class Reporter(InterfaceBase, AbstractContextManager, SetupUploadMixin, AsyncMan
         :type value: int
         :param path: A path to an image file. Required unless matrix is provided.
         :type path: str
-        :param matrix: A 3D numpy.ndarray object containing image data (BGR). Required unless filename is provided.
+        :param matrix: A 3D numpy.ndarray object containing image data (RGB). Required unless filename is provided.
         :type matrix: str
         :param max_image_history: maximum number of image to store per metric/variant combination
         use negative value for unlimited. default is set in global configuration (default=5)

trains/binding/frameworks/tensorflow_bind.py

@@ -4,11 +4,12 @@ import sys
 import threading
 from collections import defaultdict
 from functools import partial
+from io import BytesIO
 from logging import ERROR, WARNING, getLogger
 from typing import Any
 
-import cv2
 import numpy as np
+from PIL import Image
 
 from ..frameworks import _patched_call, WeightsFileHandler, _Empty, TrainsFrameworkAdapter
 from ..import_bind import PostImportHookPatching
@@ -163,8 +164,11 @@ class EventTrainsWriter(object):
     def _decode_image(self, img_str, width, height, color_channels):
         # noinspection PyBroadException
         try:
-            image_string = np.asarray(bytearray(base64.b64decode(img_str)), dtype=np.uint8)
+            imdata = base64.b64decode(img_str)
-            image = cv2.imdecode(image_string, cv2.IMREAD_COLOR)
+            output = BytesIO(imdata)
+            im = Image.open(output)
+            image = np.asarray(im)
+            output.close()
             val = image.reshape(height, width, -1).astype(np.uint8)
             if val.ndim == 3 and val.shape[2] == 3:
                 if self._visualization_mode == 'BGR':

trains/binding/matplotlib_bind.py

@@ -6,6 +6,7 @@ from tempfile import mkstemp
 
 import six
 from six import BytesIO
+import threading
 
 from ..debugging.log import LoggerRoot
 from ..config import running_remotely
@@ -21,6 +22,7 @@ class PatchedMatplotlib:
     _global_image_counter = -1
     _current_task = None
     _support_image_plot = False
+    _recursion_guard = {}
 
     class _PatchWarnings(object):
         def __init__(self):
@@ -115,19 +117,21 @@ class PatchedMatplotlib:
 
     @staticmethod
     def patched_figure_show(self, *args, **kw):
-        if hasattr(self, '_trains_show'):
+        tid = threading._get_ident() if six.PY2 else threading.get_ident()
-            # flag will be cleared when calling clf() (object will be replaced)
+        if PatchedMatplotlib._recursion_guard.get(tid):
+            # we are inside a gaurd do nothing
             return PatchedMatplotlib._patched_original_figure(self, *args, **kw)
-        try:
+
-            self._trains_show = True
+        PatchedMatplotlib._recursion_guard[tid] = True
-        except Exception:
-            pass
         PatchedMatplotlib._report_figure(set_active=False, specific_fig=self)
         ret = PatchedMatplotlib._patched_original_figure(self, *args, **kw)
+        PatchedMatplotlib._recursion_guard[tid] = False
         return ret
 
     @staticmethod
     def patched_show(*args, **kw):
+        tid = threading._get_ident() if six.PY2 else threading.get_ident()
+        PatchedMatplotlib._recursion_guard[tid] = True
         # noinspection PyBroadException
         try:
             figures = PatchedMatplotlib._get_output_figures(None, all_figures=True)
@@ -145,6 +149,7 @@ class PatchedMatplotlib:
                     plt.clf()
             except Exception:
                 pass
+        PatchedMatplotlib._recursion_guard[tid] = False
         return ret
 
     @staticmethod
@@ -172,12 +177,6 @@ class PatchedMatplotlib:
             else:
                 mpl_fig = specific_fig
 
-            # mark as processed, so nested calls to figure.show will do nothing
-            try:
-                mpl_fig._trains_show = True
-            except Exception:
-                pass
-
             # convert to plotly
             image = None
             plotly_fig = None

trains/utilities/seed.py

@@ -5,11 +5,6 @@ try:
     import numpy as np
 except Exception:
     np = None
-try:
-    import cv2
-except Exception:
-    cv2 = None
-
 
 def make_deterministic(seed=1337, cudnn_deterministic=False):
     """
@@ -40,12 +35,6 @@ def make_deterministic(seed=1337, cudnn_deterministic=False):
     if np is not None:
         np.random.seed(seed)
 
-    if cv2 is not None:
-        try:
-            cv2.setRNGSeed(seed)
-        except Exception:
-            pass
-
     if torch is not None:
         try:
             torch.manual_seed(seed)