update issue template; remove unused "set_shape" in MaskRCNN

.github/ISSUE_TEMPLATE.md

@@ -20,13 +20,13 @@ Some typical questions that we DO NOT answer:
   We have no plans to do so. We don't consider feature
   requests for examples or implement a paper for you.
   If you don't know how to do something yourself, you may ask a usage question.
-+ "The examples do not perform well after I change the models/dataset/parameters/etc."
-  Tensorpack maintainers make sure the examples perform well without modification.
++ "The examples do not perform as expected after I change the models/dataset/parameters/etc."
+  Tensorpack maintainers make sure the examples perform well without modifications.
   But it's your job to pick the model and parameters that are suitable for your own situation.
   We cannot help with such questions unless they appear to be a bug in tensorpack.
-+ "Why my model doesn't work?", "I don't understand this paper you implement.",
-  "How should I change the examples for my own dataset?"
++ "Why my own model doesn't perform well?", "I don't understand this paper you implement.",
+  "How should I change the parameters for my own dataset?"
   We do not answer machine learning questions.
-  
+
 
 You can also use gitter (https://gitter.im/tensorpack/users) for more casual discussions.

.github/ISSUE_TEMPLATE/using-understanding-tensorpack.md

@@ -9,21 +9,28 @@ about: More general questions about Tensorpack.
 
 + Your question is probably answered in [tutorials](http://tensorpack.readthedocs.io/en/latest/tutorial/index.html#user-tutorials). Read it first.
 
-+ We answer "HOW to do X with Tensorpack" for a well-defined X.
-  We also answer "HOW/WHY Tensorpack does X" for some X that Tensorpack or its examples are doing.
++ We answer "HOW to do X with Tensorpack" for a well-defined specific X.
+  X must be something that you conceptually know how to do, but are unable to do due to lack of knowledge about Tensorpack.
+  We don't answer vague questions like "How to do semantic segmentation in Tensorpack".
+
++ We also answer "HOW/WHY Tensorpack does X" for some X that Tensorpack or its examples are doing.
+
 
 Some typical questions that we DO NOT answer:
 
-+ "Could you improve/implement an example/paper ?" --
++ "Could you improve/implement an example/paper?":
+
   We have no plans to do so. We don't consider feature
   requests for examples or implement a paper for you.
   If you don't know how to do something yourself, you may ask a usage question.
-+ "The examples do not perform well after I change the models/dataset/parameters/etc."
-  Tensorpack maintainers make sure the examples perform well without modification.
++ "The examples do not perform as expected after I change the models/dataset/parameters/etc.":
+
+  Tensorpack maintainers make sure the examples perform well without modifications.
   But it's your job to pick the model and parameters that are suitable for your own situation.
   We cannot help with such questions unless they appear to be a bug in tensorpack.
-+ "Why my model doesn't work?", "I don't understand this paper you implement.",
-  "How should I change the examples for my own dataset?"
-  We do not answer machine learning questions. 
++ "Why my own model doesn't perform well?", "I don't understand this paper you implement.",
+  "How should I change the parameters for my own dataset?":
+
+  We do not answer machine learning questions.
 
 You can also use gitter (https://gitter.im/tensorpack/users) for more casual discussions.

examples/FasterRCNN/model_frcnn.py

@@ -190,7 +190,6 @@ def fastrcnn_predictions(boxes, scores):
     assert boxes.shape[1] == cfg.DATA.NUM_CLASS
     assert scores.shape[1] == cfg.DATA.NUM_CLASS
     boxes = tf.transpose(boxes, [1, 0, 2])[1:, :, :]  # #catxnx4
-    boxes.set_shape([None, cfg.DATA.NUM_CATEGORY, None])
     scores = tf.transpose(scores[:, 1:], [1, 0])  # #catxn
 
     def f(X):

examples/README.md

@@ -5,7 +5,8 @@ Training examples with __reproducible performance__.
 
 __The word "reproduce" should always mean reproduce performance__.
 With the magic of SGD, wrong deep learning code often appears to work, especially if you try it on toy datasets.
-Github is full of such deep learning code that "implements" but does not "reproduce" methods.
+Github is full of deep learning code that "implements" but does not "reproduce"
+methods, and you'll not know whether the implementation is actually correct.
 See [Unawareness of Deep Learning Mistakes](https://medium.com/@ppwwyyxx/unawareness-of-deep-learning-mistakes-d5b5774da0ba).
 
 We refuse toy examples.