Merge pull request #211 from microsoft/master

merge master

src/sdk/pynni/nni/tuner.py

@@ -27,7 +27,6 @@ _logger = logging.getLogger(__name__)
 
 
 class Tuner(Recoverable):
-    # pylint: disable=no-self-use,unused-argument
 
     def generate_parameters(self, parameter_id, **kwargs):
         """Returns a set of trial (hyper-)parameters, as a serializable object.
@@ -47,7 +46,7 @@ class Tuner(Recoverable):
         result = []
         for parameter_id in parameter_id_list:
             try:
-                _logger.debug("generating param for {}".format(parameter_id))
+                _logger.debug("generating param for %s", parameter_id)
                 res = self.generate_parameters(parameter_id, **kwargs)
             except nni.NoMoreTrialError:
                 return result
@@ -71,6 +70,8 @@ class Tuner(Recoverable):
         By default `receive_trial_result()` will only receive results of algorithm-generated hyper-parameters.
         If tuners want to receive those of customized parameters as well, they can call this function in `__init__()`.
         """
+        # pylint: disable=attribute-defined-outside-init
+        # FIXME: because tuner is designed as interface, this API should not be here
         self._accept_customized = accept
 
     def trial_end(self, parameter_id, success, **kwargs):
@@ -78,7 +79,6 @@ class Tuner(Recoverable):
         parameter_id: int
         success: True if the trial successfully completed; False if failed or terminated
         """
-        pass
 
     def update_search_space(self, search_space):
         """Update the search space of tuner. Must override.
@@ -91,20 +91,19 @@ class Tuner(Recoverable):
         path: checkpoint directory for tuner
         """
         checkpoin_path = self.get_checkpoint_path()
-        _logger.info('Load checkpoint ignored by tuner, checkpoint path: %s' % checkpoin_path)
+        _logger.info('Load checkpoint ignored by tuner, checkpoint path: %s', checkpoin_path)
 
     def save_checkpoint(self):
         """Save the checkpoint of tuner.
         path: checkpoint directory for tuner
         """
         checkpoin_path = self.get_checkpoint_path()
-        _logger.info('Save checkpoint ignored by tuner, checkpoint path: %s' % checkpoin_path)
+        _logger.info('Save checkpoint ignored by tuner, checkpoint path: %s', checkpoin_path)
 
     def import_data(self, data):
         """Import additional data for tuning
         data: a list of dictionarys, each of which has at least two keys, 'parameter' and 'value'
         """
-        pass
 
     def _on_exit(self):
         pass

src/sdk/pynni/nni/utils.py

@@ -84,12 +84,13 @@ def extract_scalar_reward(value, scalar_key='default'):
         Incorrect final result: the final result should be float/int,
         or a dict which has a key named "default" whose value is float/int.
     """
-    if isinstance(value, float) or isinstance(value, int):
+    if isinstance(value, (float, int)):
         reward = value
     elif isinstance(value, dict) and scalar_key in value and isinstance(value[scalar_key], (float, int)):
         reward = value[scalar_key]
     else:
-        raise RuntimeError('Incorrect final result: the final result should be float/int, or a dict which has a key named "default" whose value is float/int.')
+        raise RuntimeError('Incorrect final result: the final result should be float/int, ' \
+            'or a dict which has a key named "default" whose value is float/int.')
     return reward
 
 
@@ -101,7 +102,6 @@ def convert_dict2tuple(value):
         for _keys in value:
             value[_keys] = convert_dict2tuple(value[_keys])
         return tuple(sorted(value.items()))
-    else:
     return value
 
 

src/sdk/pynni/pylintrc

+# This pylintrc file is a little more strick than the one in root of code directory
+# SDK source MUST pass lint rules in top level directory, and SHOULD pass rules here
+
+[SETTINGS]
+
+max-line-length=140
+
+disable =
+    missing-docstring,
+    invalid-name,                   # C0103
+    no-member,                      # E1101: sometimes pylint cannot detect members correctly due to a bug
+    c-extension-no-member,          # I1101
+    no-self-use,                    # R0201: many functions in this SDK are designed for override
+    duplicate-code,                 # R0801
+    too-many-instance-attributes,   # R0902
+    too-few-public-methods,         # R0903
+    too-many-public-methods,        # R0904
+    too-many-return-statements,     # R0911
+    too-many-branches,              # R0912
+    too-many-arguments,             # R0913
+    too-many-locals,                # R0914
+    too-many-statements,            # R0915
+    too-many-nested-blocks,         # R1702
+    no-else-return,                 # R1705
+    chained-comparison,             # R1716
+    no-else-raise,                  # R1720
+    protected-access,               # W0212: underscore variables may be protected by whole SDK instead of single module
+    arguments-differ,               # W0221: pylint cannot handle *args and **kwargs
+    super-init-not-called,          # W0231: some interface classes do not expect users to call init
+    useless-super-delegation,       # W0235: derived init may have different docstring
+    global-statement,               # W0603: globals are useful to hide SDK internal states from user
+    unused-argument,                # W0613: many functions in this SDK are designed for override
+    broad-except,                   # W0703: the SDK commonly catch exceptions to report error
+    fixme,                          # W0511
+
+ignore-patterns=test.*.py
+
+# List of members which are set dynamically and missed by pylint inference
+generated-members=numpy.*,torch.*

src/sdk/pynni/tests/test_compressor.py

@@ -100,21 +100,21 @@ class TorchMnist(torch.nn.Module):
 class CompressorTestCase(TestCase):
     def test_tf_pruner(self):
         model = TfMnist()
-        configure_list = [{'sparsity': 0.8, 'op_types': 'default'}]
+        configure_list = [{'sparsity': 0.8, 'op_types': ['default']}]
         tf_compressor.LevelPruner(configure_list).compress_default_graph()
 
     def test_tf_quantizer(self):
         model = TfMnist()
-        tf_compressor.NaiveQuantizer([{'op_types': 'default'}]).compress_default_graph()
+        tf_compressor.NaiveQuantizer([{'op_types': ['default']}]).compress_default_graph()
 
     def test_torch_pruner(self):
         model = TorchMnist()
-        configure_list = [{'sparsity': 0.8, 'op_types': 'default'}]
+        configure_list = [{'sparsity': 0.8, 'op_types': ['default']}]
         torch_compressor.LevelPruner(configure_list).compress(model)
 
     def test_torch_quantizer(self):
         model = TorchMnist()
-        torch_compressor.NaiveQuantizer([{'op_types': 'default'}]).compress(model)
+        torch_compressor.NaiveQuantizer([{'op_types': ['default']}]).compress(model)
 
 
 if __name__ == '__main__':

src/webui/src/App.tsx

@@ -33,7 +33,7 @@ class App extends React.Component<{}, AppState> {
     }
 
     changeInterval = (interval: number) => {
-        this.setState({ interval: interval });
+        this.setState({ interval });
         if (this.timerId === null && interval !== 0) {
             window.setTimeout(this.refresh);
         } else if (this.timerId !== null && interval === 0) {

tools/nni_annotation/__init__.py

@@ -154,4 +154,3 @@ def _generate_specific_file(src_path, dst_path, exp_id, trial_id, module):
                 raise RuntimeError(src_path + ' ' + '\n'.join(str(arg) for arg in exc.args))
             else:
                 raise RuntimeError('Failed to expand annotations for %s: %r' % (src_path, exc))
-

tools/nni_annotation/search_space_generator.py

@@ -20,9 +20,10 @@
 
 
 import ast
-import astor
 import numbers
 
+import astor
+
 # pylint: disable=unidiomatic-typecheck