Merge pull request #94 from bnsh/master

Added the ability to handle multiple embeddings per epoch.

demo_multiple_embedding.py

@@ -0,0 +1,50 @@
+import math
+import torch
+from tensorboardX import SummaryWriter
+
+def main():
+    degrees = torch.arange(3600).resize_(3600, 1) * math.pi / 180.0
+    labels = ["%d" % (i) for i in range(0, 3600)]
+
+    with SummaryWriter() as writer:
+        # Maybe make a bunch of data that's always shifted in some
+        # way, and that will be hard for PCA to turn into a sphere?
+
+        for epoch in range(0, 16):
+            shift = epoch * 2 * math.pi / 16.0
+            mat = torch.cat([ \
+                torch.sin(shift + degrees * 2 * math.pi / 180.0), \
+                torch.sin(shift + degrees * 3 * math.pi / 180.0), \
+                torch.sin(shift + degrees * 5 * math.pi / 180.0), \
+                torch.sin(shift + degrees * 7 * math.pi / 180.0), \
+                torch.sin(shift + degrees * 11 * math.pi / 180.0) \
+            ], dim=1)
+            writer.add_embedding(mat=mat, metadata=labels, tag="sin", global_step=epoch)
+
+            mat = torch.cat([ \
+                torch.cos(shift + degrees * 2 * math.pi / 180.0), \
+                torch.cos(shift + degrees * 3 * math.pi / 180.0), \
+                torch.cos(shift + degrees * 5 * math.pi / 180.0), \
+                torch.cos(shift + degrees * 7 * math.pi / 180.0), \
+                torch.cos(shift + degrees * 11 * math.pi / 180.0) \
+            ], dim=1)
+            writer.add_embedding(mat=mat, metadata=labels, tag="cos", global_step=epoch)
+
+            mat = torch.cat([ \
+                torch.tan(shift + degrees * 2 * math.pi / 180.0), \
+                torch.tan(shift + degrees * 3 * math.pi / 180.0), \
+                torch.tan(shift + degrees * 5 * math.pi / 180.0), \
+                torch.tan(shift + degrees * 7 * math.pi / 180.0), \
+                torch.tan(shift + degrees * 11 * math.pi / 180.0) \
+            ], dim=1)
+            writer.add_embedding(mat=mat, metadata=labels, tag="tan", global_step=epoch)
+
+if __name__ == "__main__":
+    main()
+
+# tensorboard --logdir runs
+# Under "Projection, you should see
+#  48 tensor found named
+#     cos:cos-00000 to cos:cos-00016
+#     sin:sin-00000 to sin:sin-00016
+#     tan:tan-00000 to tan:tan-00016

tensorboardX/embedding.py

@@ -25,17 +25,17 @@ def make_sprite(label_img, save_path):
     torchvision.utils.save_image(label_img, os.path.join(save_path, 'sprite.png'), nrow=nrow, padding=0)
 
 
-def append_pbtxt(metadata, label_img, save_path, global_step, tag):
+def append_pbtxt(metadata, label_img, save_path, subdir, global_step, tag):
     with open(os.path.join(save_path, 'projector_config.pbtxt'), 'a') as f:
         # step = os.path.split(save_path)[-1]
         f.write('embeddings {\n')
-        f.write('tensor_name: "{}:{}"\n'.format(tag, global_step))
-        f.write('tensor_path: "{}"\n'.format(os.path.join(global_step, 'tensors.tsv')))
+        f.write('tensor_name: "{}:{}"\n'.format(tag, str(global_step).zfill(5)))
+        f.write('tensor_path: "{}"\n'.format(os.path.join(subdir, 'tensors.tsv')))
         if metadata is not None:
-            f.write('metadata_path: "{}"\n'.format(os.path.join(global_step, 'metadata.tsv')))
+            f.write('metadata_path: "{}"\n'.format(os.path.join(subdir, 'metadata.tsv')))
         if label_img is not None:
             f.write('sprite {\n')
-            f.write('image_path: "{}"\n'.format(os.path.join(global_step, 'sprite.png')))
+            f.write('image_path: "{}"\n'.format(os.path.join(subdir, 'sprite.png')))
             f.write('single_image_dim: {}\n'.format(label_img.size(3)))
             f.write('single_image_dim: {}\n'.format(label_img.size(2)))
             f.write('}\n')

tensorboardX/writer.py

@@ -399,6 +399,15 @@ def add_graph(self, model, input_to_model, verbose=False):
                 return
         self.file_writer.add_graph(graph(model, input_to_model, verbose))
 
+    @staticmethod
+    def _encode(rawstr):
+        # I'd use urllib but, I'm unsure about the differences from python3 to python2, etc.
+        retval = rawstr
+        retval = retval.replace("%", "%%%02x" % (ord("%")))
+        retval = retval.replace("/", "%%%02x" % (ord("/")))
+        retval = retval.replace("\\", "%%%02x" % (ord("\\")))
+        return retval
+
     def add_embedding(self, mat, metadata=None, label_img=None, global_step=None, tag='default'):
         """Add embedding projector data to summary.
 
@@ -436,7 +445,10 @@ def add_embedding(self, mat, metadata=None, label_img=None, global_step=None, ta
         if global_step is None:
             global_step = 0
             # clear pbtxt?
-        save_path = os.path.join(self.file_writer.get_logdir(), str(global_step).zfill(5))
+        # Maybe we should encode the tag so slashes don't trip us up?
+        # I don't think this will mess us up, but better safe than sorry.
+        subdir = "%s/%s" % (str(global_step).zfill(5), self._encode(tag))
+        save_path = os.path.join(self.file_writer.get_logdir(), subdir)
         try:
             os.makedirs(save_path)
         except OSError:
@@ -450,7 +462,7 @@ def add_embedding(self, mat, metadata=None, label_img=None, global_step=None, ta
         assert mat.dim() == 2, 'mat should be 2D, where mat.size(0) is the number of data points'
         make_mat(mat.tolist(), save_path)
         # new funcion to append to the config file a new embedding
-        append_pbtxt(metadata, label_img, self.file_writer.get_logdir(), str(global_step).zfill(5), tag)
+        append_pbtxt(metadata, label_img, self.file_writer.get_logdir(), subdir, global_step, tag)
 
     def add_pr_curve(self, tag, labels, predictions, global_step=None, num_thresholds=127, weights=None):
         """Adds precision recall curve.