Add name to logger.autodir for different runs and simplify (#301)

* Add name to logger.autodir for different runs and simplify
similarity learning (incl. removing tf.FLAGS in favor of argparse)

* remove 'required' from mnist-embedding.py

* next fix

* c'mon travis

* Update logger.py

* fix linting..

examples/SimilarityLearning/mnist-embeddings.py

@@ -9,7 +9,7 @@ import os
 from tensorpack import *
 import tensorpack.tfutils.symbolic_functions as symbf
 from tensorpack.tfutils.summary import add_moving_summary
-
+import argparse
 import tensorflow as tf
 import tensorflow.contrib.slim as slim
 
@@ -25,13 +25,6 @@ except ImportError:
     MATPLOTLIB_AVAIBLABLE = False
 
 
-FLAGS = tf.flags.FLAGS
-tf.flags.DEFINE_string('load', "", 'load model')
-tf.flags.DEFINE_integer('gpu', 0, 'used gpu')
-tf.flags.DEFINE_string('algorithm', "siamese", 'algorithm')
-tf.flags.DEFINE_boolean('visualize', False, 'show embedding')
-
-
 class EmbeddingModel(ModelDesc):
     def embed(self, x, nfeatures=2):
         """Embed all given tensors into an nfeatures-dim space.  """
@@ -141,9 +134,6 @@ class SoftTripletModel(TripletModel):
 
 
 def get_config(model, algorithm_name):
-    logger.set_logger_dir(
-        os.path.join('train_log',
-                     'mnist-embeddings-{}'.format(algorithm_name)))
 
     extra_display = ["cost"]
     if not algorithm_name == "cosine":
@@ -166,7 +156,7 @@ def get_config(model, algorithm_name):
     )
 
 
-def visualize(model_path, model):
+def visualize(model_path, model, algo_name):
     if not MATPLOTLIB_AVAIBLABLE:
         logger.error("visualize requires matplotlib package ...")
         return
@@ -213,27 +203,32 @@ def visualize(model_path, model):
 
     plt.axis([ax_min[0], ax_max[0], ax_min[1], ax_max[1]])
     plt.xticks([]), plt.yticks([])
-    algo_name = FLAGS.algorithm
     plt.title('Embedding using %s-loss' % algo_name)
     plt.savefig('%s.jpg' % algo_name)
 
 
 if __name__ == '__main__':
-    unknown = FLAGS._parse_flags()
-    assert len(unknown) == 0, "Invalid argument!"
-    assert FLAGS.algorithm in ["siamese", "cosine", "triplet", "softtriplet"]
+    parser = argparse.ArgumentParser()
+    parser.add_argument('--gpu', help='comma separated list of GPU(s) to use.')
+    parser.add_argument('--load', help='load model')
+    parser.add_argument('-a', '--algorithm', help='used algorithm', type=str,
+                        choices=["siamese", "cosine", "triplet", "softtriplet"])
+    parser.add_argument('--visualize', help='export embeddings into an image', action='store_true')
+    args = parser.parse_args()
 
     ALGO_CONFIGS = {"siamese": SiameseModel,
                     "cosine": CosineModel,
                     "triplet": TripletModel,
                     "softtriplet": SoftTripletModel}
 
-    with change_gpu(FLAGS.gpu):
-        if FLAGS.visualize:
-            visualize(FLAGS.load, ALGO_CONFIGS[FLAGS.algorithm])
+    logger.auto_set_dir(name=args.algorithm)
+
+    with change_gpu(args.gpu):
+        if args.visualize:
+            visualize(args.load, ALGO_CONFIGS[args.algorithm], args.algorithm)
         else:
-            config = get_config(ALGO_CONFIGS[FLAGS.algorithm], FLAGS.algorithm)
-            if FLAGS.load:
-                config.session_init = SaverRestore(FLAGS.load)
+            config = get_config(ALGO_CONFIGS[args.algorithm], args.algorithm)
+            if args.load:
+                config.session_init = SaverRestore(args.load)
             else:
                 SimpleTrainer(config).train()

tensorpack/utils/logger.py

@@ -123,13 +123,13 @@ def disable_logger():
         globals()[func] = lambda x: None
 
 
-def auto_set_dir(action=None):
+def auto_set_dir(action=None, name=None):
     """
-    Set log directory to a subdir inside "train_log", with the name being
-    the main python file currently running"""
+    Use :func:`logger.set_logger_dir` to set log directory to
+    "./train_log/{scriptname}:{name}". "scriptname" is the name of the main python file currently running"""
     mod = sys.modules['__main__']
     basename = os.path.basename(mod.__file__)
-    set_logger_dir(
-        os.path.join('train_log',
-                     basename[:basename.rfind('.')]),
-        action=action)
+    auto_dirname = os.path.join('train_log', basename[:basename.rfind('.')])
+    if name:
+        auto_dirname += ':%s' % name
+    set_logger_dir(auto_dirname, action=action)