[GitHub] marcoabreu commented on issue #9596: Remove the gradle wrapper.

[GitBox]

marcoabreu commented on issue #9596: Remove the gradle wrapper.
URL: https://github.com/apache/incubator-mxnet/pull/9596#issuecomment-360967918
 
 
   I'm afraid that this case is not covered by the Integration test. Could you 
please explain how you verified that the build is still working?


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] reminisce commented on issue #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model Quantization with Calibration

[GitBox]

reminisce commented on issue #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model 
Quantization with Calibration
URL: https://github.com/apache/incubator-mxnet/pull/9552#issuecomment-360965769
 
 
   @marcoabreu Oh, I see. Since calibration is conducted offline, it's not 
constrained by the hardware resources of edge devices. I believe there is an 
optimal value of num_bins for each layer. It could become a hyperparameter for 
users to tune.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] thinksanky opened a new pull request #9597: update copyright info

[GitBox]

thinksanky opened a new pull request #9597: update copyright info
URL: https://github.com/apache/incubator-mxnet/pull/9597
 
 
   ## Description ##
   Updated copyright information in the footer


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] marcoabreu commented on issue #9584: bump up version to 1.1

[GitBox]

marcoabreu commented on issue #9584: bump up version to 1.1
URL: https://github.com/apache/incubator-mxnet/pull/9584#issuecomment-360965685
 
 
   You might want to change the version on the website as well: 
https://github.com/apache/incubator-mxnet/blob/master/docs/_static/mxnet-theme/index.html


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[incubator-mxnet] branch master updated: Remove unnecessary RuntimeError (#9578)

[zhasheng]

This is an automated email from the ASF dual-hosted git repository.

zhasheng pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git


The following commit(s) were added to refs/heads/master by this push:
 new 45366a4  Remove unnecessary RuntimeError (#9578)
45366a4 is described below

commit 45366a49e4706b6aa0560f31f98458d8babe7d9c
Author: Anirudh Subramanian 
AuthorDate: Fri Jan 26 22:53:47 2018 -0800

Remove unnecessary RuntimeError (#9578)

* Add checksum for test_consistency

* Remove unnecessary exception raised
---
 tests/python/gpu/test_forward.py | 17 ++---
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/tests/python/gpu/test_forward.py b/tests/python/gpu/test_forward.py
index 793ad54..aae6206 100644
--- a/tests/python/gpu/test_forward.py
+++ b/tests/python/gpu/test_forward.py
@@ -44,17 +44,12 @@ def _dump_images(shape):
 def _get_data(shape):
 hash_test_img = "355e15800642286e7fe607d87c38aeeab085b0cc"
 hash_inception_v3 = "91807dfdbd336eb3b265dd62c2408882462752b9"
-fname = utils.download("http://data.mxnet.io/data/test_images_%d_%d.npy; % 
(shape),
-   path="data/test_images_%d_%d.npy" % (shape),
-   sha1_hash=hash_test_img)
-if not utils.check_sha1(fname, hash_test_img):
-raise RuntimeError("File %s not downloaded completely" % 
("test_images_%d_%d.npy"%(shape)))
-
-fname = utils.download("http://data.mxnet.io/data/inception-v3-dump.npz;,
-   path='data/inception-v3-dump.npz',
-   sha1_hash=hash_inception_v3)
-if not utils.check_sha1(fname, hash_inception_v3):
-raise RuntimeError("File %s not downloaded completely" % 
("inception-v3-dump.npz"))
+utils.download("http://data.mxnet.io/data/test_images_%d_%d.npy; % (shape),
+   path="data/test_images_%d_%d.npy" % (shape),
+   sha1_hash=hash_test_img)
+utils.download("http://data.mxnet.io/data/inception-v3-dump.npz;,
+   path='data/inception-v3-dump.npz',
+   sha1_hash=hash_inception_v3)
 
 def test_consistency(dump=False):
 shape = (299, 299)

-- 
To stop receiving notification emails like this one, please contact
zhash...@apache.org.

[GitHub] szha closed pull request #9578: Remove unnecessary RuntimeError

[GitBox]

szha closed pull request #9578: Remove unnecessary RuntimeError
URL: https://github.com/apache/incubator-mxnet/pull/9578
 
 
   

This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:

As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):

diff --git a/tests/python/gpu/test_forward.py b/tests/python/gpu/test_forward.py
index 793ad549de..aae62065a6 100644
--- a/tests/python/gpu/test_forward.py
+++ b/tests/python/gpu/test_forward.py
@@ -44,17 +44,12 @@ def _dump_images(shape):
 def _get_data(shape):
 hash_test_img = "355e15800642286e7fe607d87c38aeeab085b0cc"
 hash_inception_v3 = "91807dfdbd336eb3b265dd62c2408882462752b9"
-fname = utils.download("http://data.mxnet.io/data/test_images_%d_%d.npy; % 
(shape),
-   path="data/test_images_%d_%d.npy" % (shape),
-   sha1_hash=hash_test_img)
-if not utils.check_sha1(fname, hash_test_img):
-raise RuntimeError("File %s not downloaded completely" % 
("test_images_%d_%d.npy"%(shape)))
-
-fname = utils.download("http://data.mxnet.io/data/inception-v3-dump.npz;,
-   path='data/inception-v3-dump.npz',
-   sha1_hash=hash_inception_v3)
-if not utils.check_sha1(fname, hash_inception_v3):
-raise RuntimeError("File %s not downloaded completely" % 
("inception-v3-dump.npz"))
+utils.download("http://data.mxnet.io/data/test_images_%d_%d.npy; % (shape),
+   path="data/test_images_%d_%d.npy" % (shape),
+   sha1_hash=hash_test_img)
+utils.download("http://data.mxnet.io/data/inception-v3-dump.npz;,
+   path='data/inception-v3-dump.npz',
+   sha1_hash=hash_inception_v3)
 
 def test_consistency(dump=False):
 shape = (299, 299)


 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] marcoabreu commented on issue #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model Quantization with Calibration

[GitBox]

marcoabreu commented on issue #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model 
Quantization with Calibration
URL: https://github.com/apache/incubator-mxnet/pull/9552#issuecomment-360965066
 
 
   Oh sorry, this question was targeted towards the ``num_bins`` question - the 
email messed it up. 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] reminisce commented on issue #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model Quantization with Calibration

[GitBox]

reminisce commented on issue #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model 
Quantization with Calibration
URL: https://github.com/apache/incubator-mxnet/pull/9552#issuecomment-360964981
 
 
   @marcoabreu The optimal values are determined by the calibration datasets. 
So they are independent of platforms. So long as the platform supports int8 
basic addition and multiplication, it would be able to run quantized models. We 
would of course need to write dedicated int8 operators for a specific platform. 
The current implementation only works on Nvidia GPUs with dp4a instruction.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] reminisce commented on issue #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model Quantization with Calibration

[GitBox]

reminisce commented on issue #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model 
Quantization with Calibration
URL: https://github.com/apache/incubator-mxnet/pull/9552#issuecomment-360964778
 
 
   @jinhuang415 
   1. The parameters are quantized offline, which means the min/max values were 
pre-calculated before inference.
   2. In theory, if the calibration dataset is representative enough of the 
real inference image sets, more examples used for calibration should lead to 
less accuracy loss. The purpose of using entropy calibration is to keep the 
accuracy loss stable with respect to the number of examples used for 
calibration. The naive calibration approach suffers from more calibration 
examples leads to bigger accuracy loss as you can see the trend in the last two 
tables. My guess is that if the calibration dataset contains examples that are 
not similar to real inference images, the quantization thresholds might be 
biased by those examples and result in a little drop down of accuracy.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] indhub opened a new pull request #9596: Remove the gradle wrapper.

[GitBox]

indhub opened a new pull request #9596: Remove the gradle wrapper.
URL: https://github.com/apache/incubator-mxnet/pull/9596
 
 
   ## Description ##
   Remove the gradle wrapper. Apache RAT complains about the jar file being 
present in repo. Users can build using gradle directly.
   
   ## Checklist ##
   ### Essentials ###
   - [x] Passed code style checking (`make lint`)
   - [x] Changes are complete (i.e. I finished coding on this PR)
   - [x] All changes have test coverage:
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] thinksanky opened a new pull request #9595: fixed mail:to and the install typo

[GitBox]

thinksanky opened a new pull request #9595: fixed mail:to and the install typo
URL: https://github.com/apache/incubator-mxnet/pull/9595
 
 
   ## Description ##
   Fixed mail:to and the install link typo


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] pengzhao-intel commented on issue #9545: Profiling discussion

[GitBox]

pengzhao-intel commented on issue #9545: Profiling discussion
URL: 
https://github.com/apache/incubator-mxnet/issues/9545#issuecomment-360964261
 
 
   Yes, it's very important! 
   From my experience of Theano profiling, the overhead is very small and maybe 
you can take a look.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] reminisce commented on a change in pull request #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model Quantization with Calibration

[GitBox]

reminisce commented on a change in pull request #9552: [REQUEST FOR REVIEW | DO 
NOT MERGE] Model Quantization with Calibration
URL: https://github.com/apache/incubator-mxnet/pull/9552#discussion_r164265094
 
 

 ##
 File path: python/mxnet/quantization.py
 ##
 @@ -0,0 +1,467 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.
+
+from __future__ import absolute_import
+
+try:
+from scipy import stats
+except ImportError:
+stats = None
+
+import numpy as np
+import ctypes
+import logging
+import os
+from .base import _LIB, check_call
+from .base import c_array, c_str, mx_uint, c_str_array
+from .base import NDArrayHandle, SymbolHandle
+from .symbol import Symbol, load
+from . import ndarray as nd
+from .ndarray import NDArray
+from .io import DataIter
+from .context import cpu, Context
+from .module import Module
+
+
+def _quantize_params(qsym, params):
+"""Given a quantized symbol and a dict of params that have not been 
quantized, generate quantized params.
+Currently only supports quantizing the arg_params with names of `weight` 
or `bias`, not aux_params.
+If `qsym` contains symbols that are excluded from being quantized, their 
corresponding params will
+not be quantized, but saved together with quantized params of the symbols 
that have been quantized.
+
+Parameters
+--
+qsym : Symbol
+Quantized symbol from FP32 symbol.
+params : dict of str->NDArray
+"""
+inputs_name = qsym.list_arguments()
+quantized_params = {}
+for name in inputs_name:
+if name.endswith(('weight_quantize', 'bias_quantize')):
+original_name = name[:-len('_quantize')]
+param = params[original_name]
+val, vmin, vmax = nd.contrib.quantize(data=param, 
min_range=nd.min(param),
+  max_range=nd.max(param), 
out_type='int8')
+quantized_params[name] = val
+quantized_params[name+'_min'] = vmin
+quantized_params[name+'_max'] = vmax
+elif name in params:
+quantized_params[name] = params[name]
+return quantized_params
+
+
+def _quantize_symbol(sym, excluded_symbols=None, offline_params=None):
+"""Given a symbol object representing a neural network of data type FP32, 
quantize it into a INT8 network.
+
+Parameters
+--
+sym : Symbol
+FP32 neural network symbol.
+excluded_symbols : list of symbols
+Nodes in the network that users do not want to replace with a symbol 
of INT8 data type.
+offline_params : list of strs
+Names of the parameters that users want to quantize offline. It's 
always recommended to quantize parameters
+offline so that quantizing parameters during the inference can be 
avoided.
+"""
+num_excluded_symbols = 0
+excluded_handles = []
+if excluded_symbols is not None:
+assert isinstance(excluded_symbols, list)
+num_excluded_symbols = len(excluded_symbols)
+for s in excluded_symbols:
+excluded_handles.append(s.handle)
+
+num_offline = 0
+offline = []
+if offline_params is not None:
+num_offline = len(offline_params)
+for k in offline_params:
+offline.append(c_str(k))
+
+out = SymbolHandle()
+check_call(_LIB.MXQuantizeSymbol(sym.handle,
+ ctypes.byref(out),
+ mx_uint(num_excluded_symbols),
+ c_array(SymbolHandle, excluded_handles),
+ mx_uint(num_offline),
+ c_array(ctypes.c_char_p, offline)))
+return Symbol(out)
+
+
+class _LayerOutputCollector(object):
+"""Saves layer output NDArray in a dict with layer names as keys and lists 
of NDArrays as values.
+The collected NDArrays will be used for calculating the optimal thresholds 
for quantization using
+KL divergence."""
+def __init__(self, include_layer=None, logger=None):
+self.nd_dict = {}
+self.include_layer = include_layer
+self.logger = logger
+
+def collect(self, name, ndarray):
+if self.include_layer is not 

[GitHub] zhreshold commented on a change in pull request #8918: Added in Large-Batch SGD with a warmup, and a LARS startegy. Also add?

[GitBox]

zhreshold commented on a change in pull request #8918: Added in Large-Batch SGD 
with a warmup, and a LARS startegy. Also add?
URL: https://github.com/apache/incubator-mxnet/pull/8918#discussion_r164265033
 
 

 ##
 File path: python/mxnet/lr_scheduler.py
 ##
 @@ -136,3 +136,38 @@ def __call__(self, num_update):
 else:
 return self.base_lr
 return self.base_lr
+
+class PolyScheduler(LRScheduler):
+""" Reduce the learning rate by given a list of steps.
+
+Calculate the new learning rate by::
+
+   base_lr * (1-nup/max_nup)^pwr
+   if nup < max_nup, 0 otherwise.
+
+Parameters
+--
+   num_update: current number of updates
+   max_update: maximum number of updates before the decay reaches 0.
+   base_lr:base learning rate
+   pwr:   power of the decay term as a funtion of the current number of 
updates.
+
+"""
+
+def __init__(self, num_update, max_update, base_lr=0.01, pwr=2):
 
 Review comment:
   `num_update` useless here


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zhreshold commented on a change in pull request #8918: Added in Large-Batch SGD with a warmup, and a LARS startegy. Also add?

[GitBox]

zhreshold commented on a change in pull request #8918: Added in Large-Batch SGD 
with a warmup, and a LARS startegy. Also add?
URL: https://github.com/apache/incubator-mxnet/pull/8918#discussion_r164265043
 
 

 ##
 File path: python/mxnet/lr_scheduler.py
 ##
 @@ -136,3 +136,38 @@ def __call__(self, num_update):
 else:
 return self.base_lr
 return self.base_lr
+
+class PolyScheduler(LRScheduler):
+""" Reduce the learning rate by given a list of steps.
+
+Calculate the new learning rate by::
+
+   base_lr * (1-nup/max_nup)^pwr
+   if nup < max_nup, 0 otherwise.
+
+Parameters
+--
+   num_update: current number of updates
+   max_update: maximum number of updates before the decay reaches 0.
+   base_lr:base learning rate
+   pwr:   power of the decay term as a funtion of the current number of 
updates.
+
+"""
+
+def __init__(self, num_update, max_update, base_lr=0.01, pwr=2):
+super(PolyScheduler, self).__init__(base_lr)
+assert isinstance(max_update, int)
+if max_update < 1:
+raise ValueError("maximum number of updates must be strictly 
positive")
+self.base_lr_orig = self.base_lr
+self.max_update = max_update
+self.power = pwr
+self.count = num_update
+self.base_lr = self.base_lr_orig
+
+def __call__(self, num_update):
+if num_update <= self.max_update:
+self.base_lr = self.base_lr_orig * pow(1.0 - float(num_update) / 
float(self.max_update),
+   self.power)
+self.count += 1
 
 Review comment:
   and here


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zhreshold commented on a change in pull request #8918: Added in Large-Batch SGD with a warmup, and a LARS startegy. Also add?

[GitBox]

zhreshold commented on a change in pull request #8918: Added in Large-Batch SGD 
with a warmup, and a LARS startegy. Also add?
URL: https://github.com/apache/incubator-mxnet/pull/8918#discussion_r164265038
 
 

 ##
 File path: python/mxnet/lr_scheduler.py
 ##
 @@ -136,3 +136,38 @@ def __call__(self, num_update):
 else:
 return self.base_lr
 return self.base_lr
+
+class PolyScheduler(LRScheduler):
+""" Reduce the learning rate by given a list of steps.
+
+Calculate the new learning rate by::
+
+   base_lr * (1-nup/max_nup)^pwr
+   if nup < max_nup, 0 otherwise.
+
+Parameters
+--
+   num_update: current number of updates
+   max_update: maximum number of updates before the decay reaches 0.
+   base_lr:base learning rate
+   pwr:   power of the decay term as a funtion of the current number of 
updates.
+
+"""
+
+def __init__(self, num_update, max_update, base_lr=0.01, pwr=2):
+super(PolyScheduler, self).__init__(base_lr)
+assert isinstance(max_update, int)
+if max_update < 1:
+raise ValueError("maximum number of updates must be strictly 
positive")
+self.base_lr_orig = self.base_lr
+self.max_update = max_update
+self.power = pwr
+self.count = num_update
 
 Review comment:
   same for self.count


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] reminisce commented on issue #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model Quantization with Calibration

[GitBox]

reminisce commented on issue #9552: [REQUEST FOR REVIEW | DO NOT MERGE] Model 
Quantization with Calibration
URL: https://github.com/apache/incubator-mxnet/pull/9552#issuecomment-360963839
 
 
   @wentingj The quantized ops used in the benchmarks are convolution, 
fully-connected, avg_pooling, max_pooling, and flatten. The quantize, 
dequantize, and requantize each takes up about 5-10% runtime per epoch.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zhreshold commented on issue #9361: infer_shape error for 'resnet-152'

[GitBox]

zhreshold commented on issue #9361: infer_shape error for 'resnet-152'
URL: 
https://github.com/apache/incubator-mxnet/issues/9361#issuecomment-360963671
 
 
   You might want to checkout mxnet model serving @kevinthesun 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zhreshold commented on issue #9572: Return a NaN when using operator ( ** ) on Windows version with GPU

[GitBox]

zhreshold commented on issue #9572: Return a NaN when using operator ( ** ) on 
Windows version with GPU
URL: 
https://github.com/apache/incubator-mxnet/issues/9572#issuecomment-360963604
 
 
   Interestingly, I found it my self on windows using GPU as well: 
https://github.com/apache/incubator-mxnet/issues/9555


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zhreshold closed issue #9571: ACCURACY IS USING NUMPY, URGENT FIX

[GitBox]

zhreshold closed issue #9571: ACCURACY IS USING NUMPY, URGENT FIX
URL: https://github.com/apache/incubator-mxnet/issues/9571
 
 
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[incubator-mxnet] branch master updated: use nd for accuracy calculation (#9583)

[zhreshold]

This is an automated email from the ASF dual-hosted git repository.

zhreshold pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git


The following commit(s) were added to refs/heads/master by this push:
 new f5f1b91  use nd for accuracy calculation (#9583)
f5f1b91 is described below

commit f5f1b91ff972ad70e9131d3cd1d7408b7684
Author: Sheng Zha 
AuthorDate: Fri Jan 26 22:06:07 2018 -0800

use nd for accuracy calculation (#9583)

* use nd for accuracy calculation

* check for context
---
 python/mxnet/metric.py | 16 ++--
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/python/mxnet/metric.py b/python/mxnet/metric.py
index 5b0780a..f1cdae2 100644
--- a/python/mxnet/metric.py
+++ b/python/mxnet/metric.py
@@ -380,23 +380,27 @@ class Accuracy(EvalMetric):
 Parameters
 --
 labels : list of `NDArray`
-The labels of the data.
+The labels of the data with class indices as values, one per 
sample.
 
 preds : list of `NDArray`
-Predicted values.
+Prediction values for samples. Each prediction value can either be 
the class index,
+or a vector of likelihoods for all classes.
 """
 check_label_shapes(labels, preds)
 
 for label, pred_label in zip(labels, preds):
 if pred_label.shape != label.shape:
 pred_label = ndarray.argmax(pred_label, axis=self.axis)
-pred_label = pred_label.asnumpy().astype('int32')
-label = label.asnumpy().astype('int32')
+pred_label = pred_label.astype('int32')
+label = label.astype('int32')
 
 check_label_shapes(label, pred_label)
 
-self.sum_metric += (pred_label.flat == label.flat).sum()
-self.num_inst += len(pred_label.flat)
+if pred_label.context != label.context:
+pred_label = pred_label.as_in_context(label.context)
+
+self.sum_metric += (pred_label.flatten() == 
label.flatten()).sum().asscalar()
+self.num_inst += numpy.prod(pred_label.shape)
 
 
 @register

-- 
To stop receiving notification emails like this one, please contact
zhresh...@apache.org.

[GitHub] zhreshold closed pull request #9583: use nd for accuracy calculation

[GitBox]

zhreshold closed pull request #9583: use nd for accuracy calculation
URL: https://github.com/apache/incubator-mxnet/pull/9583
 
 
   

This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:

As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):

diff --git a/python/mxnet/metric.py b/python/mxnet/metric.py
index 5b0780aecc..f1cdae26a2 100644
--- a/python/mxnet/metric.py
+++ b/python/mxnet/metric.py
@@ -380,23 +380,27 @@ def update(self, labels, preds):
 Parameters
 --
 labels : list of `NDArray`
-The labels of the data.
+The labels of the data with class indices as values, one per 
sample.
 
 preds : list of `NDArray`
-Predicted values.
+Prediction values for samples. Each prediction value can either be 
the class index,
+or a vector of likelihoods for all classes.
 """
 check_label_shapes(labels, preds)
 
 for label, pred_label in zip(labels, preds):
 if pred_label.shape != label.shape:
 pred_label = ndarray.argmax(pred_label, axis=self.axis)
-pred_label = pred_label.asnumpy().astype('int32')
-label = label.asnumpy().astype('int32')
+pred_label = pred_label.astype('int32')
+label = label.astype('int32')
 
 check_label_shapes(label, pred_label)
 
-self.sum_metric += (pred_label.flat == label.flat).sum()
-self.num_inst += len(pred_label.flat)
+if pred_label.context != label.context:
+pred_label = pred_label.as_in_context(label.context)
+
+self.sum_metric += (pred_label.flatten() == 
label.flatten()).sum().asscalar()
+self.num_inst += numpy.prod(pred_label.shape)
 
 
 @register


 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zhreshold closed issue #9582: Misleading calculation of mxnet.metric.Accuracy

[GitBox]

zhreshold closed issue #9582: Misleading calculation of mxnet.metric.Accuracy
URL: https://github.com/apache/incubator-mxnet/issues/9582
 
 
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on issue #9581: Force the dtype of data to be float32 for test_correlation

[GitBox]

eric-haibin-lin commented on issue #9581: Force the dtype of data to be float32 
for test_correlation
URL: https://github.com/apache/incubator-mxnet/pull/9581#issuecomment-360963176
 
 
   @marcoabreu yes this could be a common problem to all other unit test using 
np random inputs


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] marcoabreu commented on issue #9581: Force the dtype of data to be float32 for test_correlation

[GitBox]

marcoabreu commented on issue #9581: Force the dtype of data to be float32 for 
test_correlation
URL: https://github.com/apache/incubator-mxnet/pull/9581#issuecomment-360963113
 
 
   So am I right when I assume that this problem could also impact other tests
   because of a dtype mismatch? Maybe we should need some type of general
   validation?
   
   Am 26.01.2018 9:19 nachm. schrieb "Haibin Lin" :
   
   > Merged #9581 .
   >
   > ?
   > You are receiving this because you were mentioned.
   > Reply to this email directly, view it on GitHub
   > ,
   > or mute the thread
   > 

   > .
   >
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin closed pull request #9535: Tutorials index page update

[GitBox]

eric-haibin-lin closed pull request #9535: Tutorials index page update
URL: https://github.com/apache/incubator-mxnet/pull/9535
 
 
   

This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:

As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):

diff --git a/docs/_static/js/sidebar.js b/docs/_static/js/sidebar.js
index 31e1450154..1c27b7e000 100644
--- a/docs/_static/js/sidebar.js
+++ b/docs/_static/js/sidebar.js
@@ -17,7 +17,7 @@ function render_left_helper(toc) {
 $('.leftsidebar > .sphinxsidebarwrapper').append(lefttoc);
 
 addToggle('.leftsidebar');
-
+
 $('.leftsidebar li a').click(function () {
 $('.leftsidebar li a').css('color', '#337ab7');
 $(this).css('color', 'black');
@@ -77,18 +77,18 @@ function render_lefttoc() {
 /*Render contents inside page*/
 function render_righttoc() {
 var url = window.location.href, indexTrailing = 'index.html';
-
+
 var rightTocTitle = "Page Contents";
 $("div.rightsidebar > div.sphinxsidebarwrapper > h3").children().remove();
 $("div.rightsidebar > div.sphinxsidebarwrapper > h3").html(rightTocTitle);
-
+
 addToggle('.rightsidebar');
-
+
 $('.rightsidebar li a').click(function () {
 $('.rightsidebar li a').css('color', '#337ab7');
 $(this).css('color', 'black');
 });
-
+
 if (url.indexOf(indexTrailing) != -1 || isAPI) {
 $('.rightsidebar').hide();
 }
@@ -175,7 +175,7 @@ function keepExpand() {
 break;
 }
 }
-
+
 //Merge right toc into left toc for API pages since they are quite long
 if (isAPI) {
 var rootEntry = currentEntry;
@@ -234,6 +234,10 @@ $(document).ready(function () {
 $('div.sphinxsidebar').hide();
 $('.content').css('width', '100%');
 }
+if (url.indexOf('/tutorials/index.html') != -1) {
+$('div.sphinxsidebar').hide();
+$('.content').css('width', '100%');
+}
 if (showRightToc) {
 $(window).scroll(function () {
 scroll_righttoc();
@@ -251,4 +255,4 @@ $(document).ready(function () {
 if ($('div.sphinxsidebar').css('visibility') == 'hidden') 
$('.content').css('width', '100%');
 return;
 }
-});
\ No newline at end of file
+});
diff --git a/docs/_static/mxnet-theme/index.html 
b/docs/_static/mxnet-theme/index.html
index f3f98f7bdc..9dfb7d6268 100644
--- a/docs/_static/mxnet-theme/index.html
+++ b/docs/_static/mxnet-theme/index.html
@@ -17,27 +17,26 @@
 
   
 
-
 
   
 
   
-Apache MXNet 1.0 Released
-We're excited to announce the release of MXNet 1.0! Check out the 
release notes for latest updates.
-https://github.com/apache/incubator-mxnet/releases/tag/1.0.0;>Learn 
More
+Apache MXNet 1.0.1 Released
+We're excited to announce the release of MXNet 1.0.1! Check out the 
release notes for latest updates.
+https://github.com/apache/incubator-mxnet/releases/tag/1.0.1;>Learn 
More
   
   
-Introducing - Gluon
-We?re happy to introduce a new elegant, easy to use, imperative 
interface for MXNet.
+MXNet Model Server
+Apache MXNet Model Server (MMS) is a flexible and easy to use tool 
for serving deep learning models exported from MXNet or the Open Neural Network 
Exchange (ONNX).
 
-http://gluon.mxnet.io;>Learn More
+https://github.com/awslabs/mxnet-model-server/blob/master/README.md;>Learn
 More
   
   
-MXNet Joining Apache
-We?re excited to announce that MXNet has been accepted to the 
Apache Incubator.
-
-http://incubator.apache.org/projects/mxnet.html;>Learn 
More
-  
+  Introducing - Gluon
+  We?re happy to introduce a new elegant, easy to use, imperative 
interface for MXNet.
+  
+  http://gluon.mxnet.io;>Learn More
+
 
   
 
diff --git a/docs/_static/mxnet-theme/layout.html 
b/docs/_static/mxnet-theme/layout.html
index 08e305c02f..653f5d7916 100644
--- a/docs/_static/mxnet-theme/layout.html
+++ b/docs/_static/mxnet-theme/layout.html
@@ -148,7 +148,7 @@
 
 https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-icon.png;>
   
-  https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-background.png;>
+  https://raw.githubusercontent.com/dmlc/web-data/master/mxnet/image/mxnet-background-compressed.jpeg;>
 
 {%- include "navbar.html" %}
 
diff --git a/docs/_static/mxnet-theme/navbar.html 
b/docs/_static/mxnet-theme/navbar.html
index e5619f17fc..bcf87f1ca7 100644
--- a/docs/_static/mxnet-theme/navbar.html
+++ b/docs/_static/mxnet-theme/navbar.html
@@ -6,7 +6,6 @@ 
   
   
 Install
-Tutorials
 
 
   Gluon 
@@ -16,7 +15,7 @@ 
  

[GitHub] eric-haibin-lin commented on issue #9412: Flaky Tests Tracking Issue

[GitBox]

eric-haibin-lin commented on issue #9412: Flaky Tests Tracking Issue
URL: 
https://github.com/apache/incubator-mxnet/issues/9412#issuecomment-360961378
 
 
#9581  fixes #9553


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on issue #9553: test_operator_gpu.test_correlation fails

[GitBox]

eric-haibin-lin commented on issue #9553: test_operator_gpu.test_correlation 
fails
URL: 
https://github.com/apache/incubator-mxnet/issues/9553#issuecomment-360961355
 
 
   closed via #9581


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin closed issue #9553: test_operator_gpu.test_correlation fails

[GitBox]

eric-haibin-lin closed issue #9553: test_operator_gpu.test_correlation fails
URL: https://github.com/apache/incubator-mxnet/issues/9553
 
 
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin closed pull request #9581: Force the dtype of data to be float32 for test_correlation

[GitBox]

eric-haibin-lin closed pull request #9581: Force the dtype of data to be 
float32 for test_correlation
URL: https://github.com/apache/incubator-mxnet/pull/9581
 
 
   

This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:

As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):

diff --git a/tests/python/unittest/test_operator.py 
b/tests/python/unittest/test_operator.py
index d169a5455b..742d055184 100644
--- a/tests/python/unittest/test_operator.py
+++ b/tests/python/unittest/test_operator.py
@@ -2072,7 +2072,9 @@ def 
correlation_backward(out_grad,tmp1,tmp2,data1,data2,pad_size,kernel_size,str
 def 
unittest_correlation(data_shape,kernel_size,max_displacement,stride1,stride2,pad_size,is_multiply):
 
 img1 = np.random.random(data_shape)
+img1 = img1.astype(np.float32)
 img2 = np.random.random(data_shape)
+img2 = img2.astype(np.float32)
 
 net1 = 
get_correlation(img1,img2,kernel_size,max_displacement,stride1,stride2,pad_size,is_multiply)
 net2 = 
get_correlation(img1,img2,kernel_size,max_displacement,stride1,stride2,pad_size,is_multiply
 )


 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin closed pull request #9577: Disable flaky test: test_correlation

[GitBox]

eric-haibin-lin closed pull request #9577: Disable flaky test: test_correlation
URL: https://github.com/apache/incubator-mxnet/pull/9577
 
 
   

This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:

As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):

diff --git a/tests/python/unittest/test_operator.py 
b/tests/python/unittest/test_operator.py
index 56dc27c493..f3ddbd51ac 100644
--- a/tests/python/unittest/test_operator.py
+++ b/tests/python/unittest/test_operator.py
@@ -2100,7 +2100,7 @@ def 
unittest_correlation(data_shape,kernel_size,max_displacement,stride1,stride2
 assert_almost_equal(exe1.grad_dict['img1'].asnumpy(), grad1, rtol=1e-3, 
atol=1e-4)
 assert_almost_equal(exe1.grad_dict['img2'].asnumpy(), grad2, rtol=1e-3, 
atol=1e-4)
 
-
+@unittest.skip("test fails intermittently. temporarily disabled till it gets 
fixed. tracked at https://github.com/apache/incubator-mxnet/issues/9553;)
 def test_correlation():
 unittest_correlation((1,3,10,10), kernel_size = 1,max_displacement = 
4,stride1 = 1,stride2 = 1,pad_size = 4,is_multiply = False)
 unittest_correlation((5,1,15,15), kernel_size = 1,max_displacement = 
5,stride1 = 1,stride2 = 1,pad_size = 5,is_multiply = False)


 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[incubator-mxnet] branch master updated: Force the dtype of data to be float32 for test_correlation (#9581)

[haibin]

This is an automated email from the ASF dual-hosted git repository.

haibin pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git


The following commit(s) were added to refs/heads/master by this push:
 new f1ae225  Force the dtype of data to be float32 for test_correlation 
(#9581)
f1ae225 is described below

commit f1ae225b56659b743b2661c6987ff07d7c4ee40f
Author: Anirudh Subramanian 
AuthorDate: Fri Jan 26 21:19:19 2018 -0800

Force the dtype of data to be float32 for test_correlation (#9581)
---
 tests/python/unittest/test_operator.py | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/tests/python/unittest/test_operator.py 
b/tests/python/unittest/test_operator.py
index d169a54..742d055 100644
--- a/tests/python/unittest/test_operator.py
+++ b/tests/python/unittest/test_operator.py
@@ -2072,7 +2072,9 @@ def 
correlation_backward(out_grad,tmp1,tmp2,data1,data2,pad_size,kernel_size,str
 def 
unittest_correlation(data_shape,kernel_size,max_displacement,stride1,stride2,pad_size,is_multiply):
 
 img1 = np.random.random(data_shape)
+img1 = img1.astype(np.float32)
 img2 = np.random.random(data_shape)
+img2 = img2.astype(np.float32)
 
 net1 = 
get_correlation(img1,img2,kernel_size,max_displacement,stride1,stride2,pad_size,is_multiply)
 net2 = 
get_correlation(img1,img2,kernel_size,max_displacement,stride1,stride2,pad_size,is_multiply
 )

-- 
To stop receiving notification emails like this one, please contact
hai...@apache.org.

[GitHub] eric-haibin-lin commented on issue #9581: Force the dtype of data to be float32 for test_correlation

[GitBox]

eric-haibin-lin commented on issue #9581: Force the dtype of data to be float32 
for test_correlation
URL: https://github.com/apache/incubator-mxnet/pull/9581#issuecomment-360961270
 
 
   @szha  Yes, we found a np seed/state that consistently fails the test. The 
random np input is in float64 dtype while the MXNet ndarray is using float32. 
Hence the difference when comparing the result. 
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin closed pull request #9575: replaced how_to with faq

[GitBox]

eric-haibin-lin closed pull request #9575: replaced how_to with faq
URL: https://github.com/apache/incubator-mxnet/pull/9575
 
 
   

This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:

As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):

diff --git a/R-package/README.md b/R-package/README.md
index e21d6b17dc..78a6214ddc 100644
--- a/R-package/README.md
+++ b/R-package/README.md
@@ -24,7 +24,7 @@ options(repos = cran)
 install.packages("mxnet")
 ```
 
-To use the GPU version or to use it on Linux, please follow [Installation 
Guide](http://mxnet.io/get_started/install.html)
+To use the GPU version or to use it on Linux, please follow [Installation 
Guide](http://mxnet.io/install/index.html)
 
 License
 ---
diff --git a/README.md b/README.md
index 5dd5f02cbd..feff029142 100644
--- a/README.md
+++ b/README.md
@@ -36,13 +36,13 @@ What's New
 * [MKLDNN for Faster CPU Performance](./MKL_README.md)
 * [MXNet Memory Monger, Training Deeper Nets with Sublinear Memory 
Cost](https://github.com/dmlc/mxnet-memonger)
 * [Tutorial for NVidia GTC 2016](https://github.com/dmlc/mxnet-gtc-tutorial)
-* [Embedding Torch layers and functions in 
MXNet](https://mxnet.incubator.apache.org/how_to/torch.html)
+* [Embedding Torch layers and functions in 
MXNet](https://mxnet.incubator.apache.org/faq/torch.html)
 * [MXNet.js: Javascript Package for Deep Learning in Browser (without server)
 ](https://github.com/dmlc/mxnet.js/)
 * [Design Note: Design Efficient Deep Learning Data Loading 
Module](https://mxnet.incubator.apache.org/architecture/note_data_loading.html)
-* [MXNet on Mobile 
Device](https://mxnet.incubator.apache.org/how_to/smart_device.html)
-* [Distributed 
Training](https://mxnet.incubator.apache.org/how_to/multi_devices.html)
-* [Guide to Creating New Operators 
(Layers)](https://mxnet.incubator.apache.org/how_to/new_op.html)
+* [MXNet on Mobile 
Device](https://mxnet.incubator.apache.org/faq/smart_device.html)
+* [Distributed 
Training](https://mxnet.incubator.apache.org/faq/multi_devices.html)
+* [Guide to Creating New Operators 
(Layers)](https://mxnet.incubator.apache.org/faq/new_op.html)
 * [Go binding for inference](https://github.com/songtianyi/go-mxnet-predictor)
 * [Amalgamation and Go Binding for 
Predictors](https://github.com/jdeng/gomxnet/) - Outdated
 * [Large Scale Image 
Classification](https://github.com/apache/incubator-mxnet/tree/master/example/image-classification)
@@ -52,10 +52,10 @@ Contents
 * [Documentation](https://mxnet.incubator.apache.org/) and  
[Tutorials](https://mxnet.incubator.apache.org/tutorials/)
 * [Design Notes](https://mxnet.incubator.apache.org/architecture/index.html)
 * [Code Examples](https://github.com/dmlc/mxnet/tree/master/example)
-* [Installation](https://mxnet.incubator.apache.org/get_started/install.html)
+* [Installation](https://mxnet.incubator.apache.org/install/index.html)
 * [Pretrained Models](https://github.com/dmlc/mxnet-model-gallery)
 * [Contribute to 
MXNet](https://mxnet.incubator.apache.org/community/contribute.html)
-* [Frequent Asked 
Questions](https://mxnet.incubator.apache.org/how_to/faq.html)
+* [Frequent Asked Questions](https://mxnet.incubator.apache.org/faq/faq.html)
 
 Features
 
diff --git a/docs/architecture/release_note_0_9.md 
b/docs/architecture/release_note_0_9.md
index 61bad50ea6..afcc091d7c 100644
--- a/docs/architecture/release_note_0_9.md
+++ b/docs/architecture/release_note_0_9.md
@@ -4,7 +4,7 @@ Version 0.9 brings a number of important features and changes, 
including a back-
 
 ## NNVM Refactor
 
-NNVM is a library for neural network graph construction, optimization, and 
operator registration. It serves as an intermediary layer between the front-end 
(MXNet user API) and the back-end (computation on the device). After version 
0.9, MXNet fully adopts the NNVM framework. Now it's easier to create 
operators. You can also register "pass"es that process and optimizes the graph 
when `bind` is called on the symbol. For more discussion on how to create 
operators with NNVM, please refer to [How to Create New 
Operators](../how_to/new_op.md)
+NNVM is a library for neural network graph construction, optimization, and 
operator registration. It serves as an intermediary layer between the front-end 
(MXNet user API) and the back-end (computation on the device). After version 
0.9, MXNet fully adopts the NNVM framework. Now it's easier to create 
operators. You can also register "pass"es that process and optimizes the graph 
when `bind` is called on the symbol. For more discussion on how to create 
operators with NNVM, please refer to [How to Create New 
Operators](../faq/new_op.md)
 
 Other changes brought by NNVM include:
 - Backward shape inference is now supported
diff --git a/docs/community/index.md b/docs/community/index.md
index 6d3f345dc9..ab98856e30 100644

[GitHub] eric-haibin-lin commented on issue #9575: replaced how_to with faq

[GitBox]

eric-haibin-lin commented on issue #9575: replaced how_to with faq
URL: https://github.com/apache/incubator-mxnet/pull/9575#issuecomment-360957382
 
 
   Merging it for now so that we can get updated nightly test results. 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[incubator-mxnet] branch master updated: replaced how_to with faq (#9575)

[haibin]

This is an automated email from the ASF dual-hosted git repository.

haibin pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git


The following commit(s) were added to refs/heads/master by this push:
 new 50850af  replaced how_to with faq (#9575)
50850af is described below

commit 50850af1ddc39cb9bbd3535ab626a53bbe9802b3
Author: thinksanky <31976455+thinksa...@users.noreply.github.com>
AuthorDate: Fri Jan 26 19:49:15 2018 -0800

replaced how_to with faq (#9575)

* replaced how_to with faq

* fixed broken links from 197 report
---
 R-package/README.md|  2 +-
 README.md  | 12 -
 docs/architecture/release_note_0_9.md  |  2 +-
 docs/community/index.md|  4 +--
 docs/faq/env_var.md|  2 +-
 docs/faq/faq.md|  4 +--
 docs/faq/finetune.md   |  2 +-
 docs/faq/gradient_compression.md   |  2 +-
 docs/faq/multi_devices.md  | 12 -
 docs/faq/nnpack.md |  2 +-
 docs/faq/perf.md   |  4 +--
 docs/faq/s3_integration.md |  2 +-
 docs/faq/visualize_graph.md|  4 +--
 docs/install/amazonlinux_setup.md  |  4 +--
 docs/install/build_from_source.md  |  2 +-
 docs/install/centos_setup.md   |  4 +--
 docs/install/osx_setup.md  |  4 +--
 docs/install/raspbian_setup.md |  4 +--
 docs/install/tx2_setup.md  |  4 +--
 docs/install/ubuntu_setup.md   |  4 +--
 docs/install/windows_setup.md  |  2 +-
 docs/tutorials/basic/data.md   | 30 +++---
 docs/tutorials/basic/module.md |  8 +++---
 docs/tutorials/basic/ndarray.md|  2 +-
 docs/tutorials/basic/symbol.md |  8 +++---
 docs/tutorials/embedded/wine_detector.md   |  4 +--
 docs/tutorials/gluon/mnist.md  |  2 +-
 docs/tutorials/python/linear-regression.md |  6 ++---
 docs/tutorials/python/mnist.md |  4 +--
 docs/tutorials/python/predict_image.md |  2 +-
 docs/tutorials/r/ndarray.md|  2 +-
 docs/tutorials/r/symbol.md |  2 +-
 docs/tutorials/scala/char_lstm.md  |  4 +--
 docs/tutorials/scala/mnist.md  |  2 +-
 docs/tutorials/scala/mxnet_scala_on_intellij.md|  2 +-
 docs/tutorials/sparse/csr.md   |  2 +-
 docs/tutorials/sparse/row_sparse.md|  2 +-
 docs/tutorials/sparse/train.md |  6 ++---
 .../tutorials/vision/large_scale_classification.md |  4 +--
 example/caffe/README.md|  2 +-
 example/image-classification/README.md |  6 ++---
 example/recommenders/crossentropy.py   |  2 +-
 example/recommenders/randomproj.py |  2 +-
 example/rnn/bucketing/README.md|  2 +-
 example/rnn/old/README.md  |  2 +-
 example/sparse/linear_classification/README.md |  2 +-
 example/ssd/tools/caffe_converter/README.md|  2 +-
 .../AI-MXNet/lib/AI/MXNet/Gluon/Trainer.pm |  2 +-
 plugin/caffe/README.md |  2 +-
 python/mxnet/context.py|  2 +-
 scala-package/README.md|  2 +-
 setup-utils/install-mxnet-osx-python.sh|  2 +-
 src/operator/custom/custom.cc  |  2 +-
 tools/caffe_converter/README.md|  2 +-
 tools/caffe_translator/README.md   |  4 +--
 tools/caffe_translator/faq.md  |  4 +--
 56 files changed, 107 insertions(+), 109 deletions(-)

diff --git a/R-package/README.md b/R-package/README.md
index e21d6b1..78a6214 100644
--- a/R-package/README.md
+++ b/R-package/README.md
@@ -24,7 +24,7 @@ options(repos = cran)
 install.packages("mxnet")
 ```
 
-To use the GPU version or to use it on Linux, please follow [Installation 
Guide](http://mxnet.io/get_started/install.html)
+To use the GPU version or to use it on Linux, please follow [Installation 
Guide](http://mxnet.io/install/index.html)
 
 License
 ---
diff --git a/README.md b/README.md
index 5dd5f02..feff029 100644
--- a/README.md
+++ b/README.md
@@ -36,13 +36,13 @@ What's New
 * [MKLDNN for Faster CPU Performance](./MKL_README.md)
 * [MXNet Memory Monger, Training Deeper Nets with Sublinear Memory 
Cost](https://github.com/dmlc/mxnet-memonger)
 * [Tutorial for NVidia GTC 2016](https://github.com/dmlc/mxnet-gtc-tutorial)
-* [Embedding Torch layers and functions in 

[GitHub] szha commented on issue #9581: Force the dtype of data to be float32 for test_correlation

[GitBox]

szha commented on issue #9581: Force the dtype of data to be float32 for 
test_correlation
URL: https://github.com/apache/incubator-mxnet/pull/9581#issuecomment-360956732
 
 
   How is the fix verified? Do you have an idea why it started failing?


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] ZiyueHuang commented on issue #9540: refactor regression ops to nnvm interface

[GitBox]

ZiyueHuang commented on issue #9540: refactor regression ops to nnvm interface
URL: https://github.com/apache/incubator-mxnet/pull/9540#issuecomment-360955082
 
 
   Seems that I mistakenly checkout that test file... Added back.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] anirudh2290 commented on issue #9380: nd.stack causes abort for differently sized arrays

[GitBox]

anirudh2290 commented on issue #9380: nd.stack causes abort for differently 
sized arrays
URL: 
https://github.com/apache/incubator-mxnet/issues/9380#issuecomment-360953495
 
 
   Related to: #7335 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] anirudh2290 commented on issue #9567: nd.argmax cause "Kernel Died" error in Jupyter Notebook

[GitBox]

anirudh2290 commented on issue #9567: nd.argmax cause "Kernel Died" error in 
Jupyter Notebook
URL: 
https://github.com/apache/incubator-mxnet/issues/9567#issuecomment-360953466
 
 
   Related to: #7335


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] wkcn commented on issue #9511: set_lr_mult() or set_wd_mult() is invalid if not setting param_idx2name for the optimizer

[GitBox]

wkcn commented on issue #9511: set_lr_mult() or set_wd_mult() is invalid if not 
setting param_idx2name for the optimizer
URL: 
https://github.com/apache/incubator-mxnet/issues/9511#issuecomment-360952988
 
 
   I think it's difficult to set_lr_mult() in the current implementation.
   I don't know how to set the idx2name value when MXNet is running on 
Multi-GPU or Multi-Machine.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on issue #9577: Disable flaky test: test_correlation

[GitBox]

eric-haibin-lin commented on issue #9577: Disable flaky test: test_correlation
URL: https://github.com/apache/incubator-mxnet/pull/9577#issuecomment-360951864
 
 
   yes
   
   On 2018?1?26? ?? at 17:28 Yizhi Liu  wrote:
   
   > I guess #9581  fixes
   > the test?
   >
   > ?
   > You are receiving this because you authored the thread.
   > Reply to this email directly, view it on GitHub
   > 
,
   > or mute the thread
   > 

   > .
   >
   -- 
   Best Regards,
   Haibin Lin
   
   Department of Computer Science
   School of Computer Science
   Carnegie Mellon University
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] ashokei commented on a change in pull request #8918: Added in Large-Batch SGD with a warmup, and a LARS startegy. Also add?

[GitBox]

ashokei commented on a change in pull request #8918: Added in Large-Batch SGD 
with a warmup, and a LARS startegy. Also add?
URL: https://github.com/apache/incubator-mxnet/pull/8918#discussion_r164259824
 
 

 ##
 File path: python/mxnet/lr_scheduler.py
 ##
 @@ -136,3 +136,42 @@ def __call__(self, num_update):
 else:
 return self.base_lr
 return self.base_lr
+
+class PolyScheduler(LRScheduler):
+""" Reduce the learning rate by given a list of steps.
+
+Calculate the new learning rate by::
+
+   base_lr * (1-nup/max_nup)^pwr
+   if nup < max_nup, 0 otherwise.
+
+Parameters
+--
+   num_update: current number of updates
+   max_update: maximum number of updates before the decay reaches 0.
+   base_lr:base learning rate
+   pwr:   power of the decay term as a funtion of the current number of 
updates.
+
+"""
+
+def __init__(self, num_update, max_update, base_lr=0.01, pwr=2):
+super(PolyScheduler, self).__init__(base_lr)
+assert isinstance(max_update, int)
+if max_update < 1:
+raise ValueError("maximum number of updates must be strictly 
positive")
+self.base_lr_orig = self.base_lr
+self.max_update = max_update
+self.power = pwr
+self.count = num_update
+if num_update <= max_update:
 
 Review comment:
   @zhreshold i removed the num_update duplicate line, can you please 
check.thanks.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on a change in pull request #9575: replaced how_to with faq

[GitBox]

eric-haibin-lin commented on a change in pull request #9575: replaced how_to 
with faq
URL: https://github.com/apache/incubator-mxnet/pull/9575#discussion_r164259422
 
 

 ##
 File path: docs/community/index.md
 ##
 @@ -8,9 +8,9 @@ We track bugs and new feature requests in the MXNet Github 
repo in the issues fo
 ## Contributors
 MXNet has been developed and is used by a group of active community members. 
Contribute to improving it! For more information, see 
[contributions](http://mxnet.io/community/contribute.html).
 
-Please join the contributor mailing list. 
[subscribe]('mailto:dev-subscr...@mxnet.incubator.apache.org') 
[archive](https://lists.apache.org/list.html?d...@mxnet.apache.org)
+Please join the contributor mailing list. 
[subscribe](mailto://dev-subscr...@mxnet.incubator.apache.org) 
[archive](https://lists.apache.org/list.html?d...@mxnet.apache.org)
 
 Review comment:
   We need to validate this on the website


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on a change in pull request #9575: replaced how_to with faq

[GitBox]

eric-haibin-lin commented on a change in pull request #9575: replaced how_to 
with faq
URL: https://github.com/apache/incubator-mxnet/pull/9575#discussion_r164259265
 
 

 ##
 File path: docs/tutorials/embedded/wine_detector.md
 ##
 @@ -37,9 +37,7 @@ To complete this tutorial, you need:
 
 ## Building MXNet for The Pi
 
-The first step will be to get MXNet with the Python bindings running on your 
Raspberry Pi 3. There is a tutorial for that provided on 
[here](http://mxnet.io/get_started/raspbian_setup.html). In short you will have 
to download the dependencies, and build the full MXNet library for the Pi with 
the ARM specific compile flags. Be sure to build the library with open CV as we 
will be using a model that requires it to process images. Then you will finally 
the Python bindings. Once this is done you should test that works by opening a 
python REPL on your Pi and typing the following commands:
-
-The first step is to get MXNet with the Python bindings running on your 
Raspberry Pi 3. There is a tutorial for that provided 
[here](http://mxnet.io/get_started/raspbian_setup.html). The linked tutorial 
walks you through downloading the dependencies, and building the full MXNet 
library for the Pi with the ARM specific compile flags. Be sure to build the 
library with open CV as we will be using a model that requires it to process 
images. Then you will register the Python bindings to MXNet. After this is done 
you should test that your installation works by opening a python REPL on your 
Pi and typing the following commands:
+The first step is to get MXNet with the Python bindings running on your 
Raspberry Pi 3. There is a tutorial for that provided 
[here](http://mxnet.io/insstall/index.html). The linked tutorial walks you 
through downloading the dependencies, and building the full MXNet library for 
the Pi with the ARM specific compile flags. Be sure to build the library with 
open CV as we will be using a model that requires it to process images. Then 
you will register the Python bindings to MXNet. After this is done you should 
test that your installation works by opening a python REPL on your Pi and 
typing the following commands:
 
 Review comment:
   `insstall ` -> `install `


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha opened a new issue #9594: mx.metric.PearsonCorrelation is using numpy logic

[GitBox]

szha opened a new issue #9594: mx.metric.PearsonCorrelation is using numpy logic
URL: https://github.com/apache/incubator-mxnet/issues/9594
 
 
   The metric module has been using numpy logic and is not benefiting from 
existing mxnet operators.
   
   
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/metric.py#L1034-L1039


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha opened a new issue #9593: mx.metric.CrossEntropy/NegativeLogLikelihood is using numpy logic and partially duplicates mx.gluon.loss.SoftmaxCELoss

[GitBox]

szha opened a new issue #9593: mx.metric.CrossEntropy/NegativeLogLikelihood is 
using numpy logic and partially duplicates mx.gluon.loss.SoftmaxCELoss
URL: https://github.com/apache/incubator-mxnet/issues/9593
 
 
   metric.CrossEntropy: 
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/metric.py#L909-L918
   metric.NegativeLogLikelihood: 
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/metric.py#L977-L986
   gluon.loss.SoftmaxCELoss: 
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/gluon/loss.py#L310-L318


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha opened a new issue #9591: mx.metric.MAE is using numpy logic and duplicates mx.gluon.loss.L1Loss

[GitBox]

szha opened a new issue #9591: mx.metric.MAE is using numpy logic and 
duplicates mx.gluon.loss.L1Loss
URL: https://github.com/apache/incubator-mxnet/issues/9591
 
 
   metric.MAE: 
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/metric.py#L725-L733
   gluon.loss.L1Loss: 
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/gluon/loss.py#L169-L173


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on a change in pull request #9583: use nd for accuracy calculation

[GitBox]

szha commented on a change in pull request #9583: use nd for accuracy 
calculation
URL: https://github.com/apache/incubator-mxnet/pull/9583#discussion_r164257186
 
 

 ##
 File path: python/mxnet/metric.py
 ##
 @@ -380,23 +380,24 @@ def update(self, labels, preds):
 Parameters
 --
 labels : list of `NDArray`
-The labels of the data.
+The labels of the data with class indices as values, one per 
sample.
 
 preds : list of `NDArray`
-Predicted values.
+Prediction values for samples. Each prediction value can either be 
the class index,
+or a vector of likelihoods for all classes.
 """
 check_label_shapes(labels, preds)
 
 for label, pred_label in zip(labels, preds):
 if pred_label.shape != label.shape:
 pred_label = ndarray.argmax(pred_label, axis=self.axis)
-pred_label = pred_label.asnumpy().astype('int32')
-label = label.asnumpy().astype('int32')
+pred_label = pred_label.astype('int32')
 
 Review comment:
   This requires larger space, which can show when the prediction class is 
large (such as in NLP applications). Should I make it an option?


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zhreshold commented on a change in pull request #9583: use nd for accuracy calculation

[GitBox]

zhreshold commented on a change in pull request #9583: use nd for accuracy 
calculation
URL: https://github.com/apache/incubator-mxnet/pull/9583#discussion_r164257671
 
 

 ##
 File path: python/mxnet/metric.py
 ##
 @@ -380,23 +380,24 @@ def update(self, labels, preds):
 Parameters
 --
 labels : list of `NDArray`
-The labels of the data.
+The labels of the data with class indices as values, one per 
sample.
 
 preds : list of `NDArray`
-Predicted values.
+Prediction values for samples. Each prediction value can either be 
the class index,
+or a vector of likelihoods for all classes.
 """
 check_label_shapes(labels, preds)
 
 for label, pred_label in zip(labels, preds):
 if pred_label.shape != label.shape:
 pred_label = ndarray.argmax(pred_label, axis=self.axis)
-pred_label = pred_label.asnumpy().astype('int32')
-label = label.asnumpy().astype('int32')
+pred_label = pred_label.astype('int32')
 
 Review comment:
   should be good in most cases.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha opened a new issue #9590: mx.metric.Perplexity should support calculating from cross-entropy

[GitBox]

szha opened a new issue #9590: mx.metric.Perplexity should support calculating 
from cross-entropy
URL: https://github.com/apache/incubator-mxnet/issues/9590
 
 
   Currently part of the logic for calculating cross-entropy loss is in 
mx.metric.Perplexity. This means we would have repeated calculation when we are 
already using cross entropy loss, which is very commonly used (e.g 
SoftmaxCrossEntropyLoss). In that case, only exponentiating is needed.
   
   
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/metric.py#L648-L658


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] yzhliu commented on issue #9577: Disable flaky test: test_correlation

[GitBox]

yzhliu commented on issue #9577: Disable flaky test: test_correlation
URL: https://github.com/apache/incubator-mxnet/pull/9577#issuecomment-360948391
 
 
   I guess https://github.com/apache/incubator-mxnet/pull/9581 fixes the test?


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on issue #9582: Misleading calculation of mxnet.metric.Accuracy

[GitBox]

szha commented on issue #9582: Misleading calculation of mxnet.metric.Accuracy
URL: 
https://github.com/apache/incubator-mxnet/issues/9582#issuecomment-360947976
 
 
   Thanks for the feedback.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on a change in pull request #9583: use nd for accuracy calculation

[GitBox]

szha commented on a change in pull request #9583: use nd for accuracy 
calculation
URL: https://github.com/apache/incubator-mxnet/pull/9583#discussion_r164257186
 
 

 ##
 File path: python/mxnet/metric.py
 ##
 @@ -380,23 +380,24 @@ def update(self, labels, preds):
 Parameters
 --
 labels : list of `NDArray`
-The labels of the data.
+The labels of the data with class indices as values, one per 
sample.
 
 preds : list of `NDArray`
-Predicted values.
+Prediction values for samples. Each prediction value can either be 
the class index,
+or a vector of likelihoods for all classes.
 """
 check_label_shapes(labels, preds)
 
 for label, pred_label in zip(labels, preds):
 if pred_label.shape != label.shape:
 pred_label = ndarray.argmax(pred_label, axis=self.axis)
-pred_label = pred_label.asnumpy().astype('int32')
-label = label.asnumpy().astype('int32')
+pred_label = pred_label.astype('int32')
 
 Review comment:
   This requires larger space can show when the prediction class is large (such 
as in NLP applications). Should I make it an option?


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zhreshold commented on a change in pull request #9583: use nd for accuracy calculation

[GitBox]

zhreshold commented on a change in pull request #9583: use nd for accuracy 
calculation
URL: https://github.com/apache/incubator-mxnet/pull/9583#discussion_r164257105
 
 

 ##
 File path: python/mxnet/metric.py
 ##
 @@ -380,23 +380,24 @@ def update(self, labels, preds):
 Parameters
 --
 labels : list of `NDArray`
-The labels of the data.
+The labels of the data with class indices as values, one per 
sample.
 
 preds : list of `NDArray`
-Predicted values.
+Prediction values for samples. Each prediction value can either be 
the class index,
+or a vector of likelihoods for all classes.
 """
 check_label_shapes(labels, preds)
 
 for label, pred_label in zip(labels, preds):
 if pred_label.shape != label.shape:
 pred_label = ndarray.argmax(pred_label, axis=self.axis)
-pred_label = pred_label.asnumpy().astype('int32')
-label = label.asnumpy().astype('int32')
+pred_label = pred_label.astype('int32')
 
 Review comment:
   `int64`is better?


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] astonzhang commented on issue #9572: Return a NaN when using operator ( ** ) on Windows version with GPU

[GitBox]

astonzhang commented on issue #9572: Return a NaN when using operator ( ** ) on 
Windows version with GPU
URL: 
https://github.com/apache/incubator-mxnet/issues/9572#issuecomment-360947529
 
 
   Related post:
   https://discuss.gluon.ai/t/topic/989
   
   This issue has been raised by multiple users. Thanks.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mbaijal commented on a change in pull request #9548: updated release version to 1.0.1 on the mainpage and re-arranged news?

[GitBox]

mbaijal commented on a change in pull request #9548: updated release version to 
1.0.1 on the mainpage and re-arranged news?
URL: https://github.com/apache/incubator-mxnet/pull/9548#discussion_r164256910
 
 

 ##
 File path: docs/_static/mxnet-theme/index.html
 ##
 @@ -17,27 +17,26 @@
 
   
 
-
 
   
 
   
-Apache MXNet 1.0 Released
-We're excited to announce the release of MXNet 1.0! Check out the 
release notes for latest updates.
-https://github.com/apache/incubator-mxnet/releases/tag/1.0.0;>Learn 
More
+Apache MXNet 1.0.1 Released
+We're excited to announce the release of MXNet 1.0.1! Check out the 
release notes for latest updates.
+https://github.com/apache/incubator-mxnet/releases/tag/1.0.1;>Learn 
More
 
 Review comment:
   Ideally, It should be `1.1.0` everywhere. Once the RC is created, no src can 
be modified so it must be the final release tag. 
   However, This would lead to a broken link on the website and I think this 
should only be merged onto the release branch for now. 
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on issue #9588: metric should have TP FP TN FN Precision Recall F1 for both macro and micro versions

[GitBox]

szha commented on issue #9588: metric should have TP FP TN FN Precision Recall 
F1 for both macro and micro versions
URL: 
https://github.com/apache/incubator-mxnet/issues/9588#issuecomment-360947519
 
 
   We may refer to the "average" option in 
[sklearn.metrics.f1_score](http://scikit-learn.org/stable/modules/generated/sklearn.metrics.f1_score.html#sklearn.metrics.f1_score).


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha opened a new issue #9589: mx.metric should support top-K version of TP FP TN FN Precision Recall F1

[GitBox]

szha opened a new issue #9589: mx.metric should support top-K version of TP FP 
TN FN Precision Recall F1
URL: https://github.com/apache/incubator-mxnet/issues/9589
 
 
   This request is the same as #9588 with the addition of expanded match logic 
based on topk prediction.
   
   https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/metric.py


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mbaijal commented on a change in pull request #9548: updated release version to 1.0.1 on the mainpage and re-arranged news?

[GitBox]

mbaijal commented on a change in pull request #9548: updated release version to 
1.0.1 on the mainpage and re-arranged news?
URL: https://github.com/apache/incubator-mxnet/pull/9548#discussion_r164256910
 
 

 ##
 File path: docs/_static/mxnet-theme/index.html
 ##
 @@ -17,27 +17,26 @@
 
   
 
-
 
   
 
   
-Apache MXNet 1.0 Released
-We're excited to announce the release of MXNet 1.0! Check out the 
release notes for latest updates.
-https://github.com/apache/incubator-mxnet/releases/tag/1.0.0;>Learn 
More
+Apache MXNet 1.0.1 Released
+We're excited to announce the release of MXNet 1.0.1! Check out the 
release notes for latest updates.
+https://github.com/apache/incubator-mxnet/releases/tag/1.0.1;>Learn 
More
 
 Review comment:
   It should be `1.1.0` everywhere. Once the RC is created, no src can be 
modified so it must be the final release tag. 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] astonzhang commented on issue #9572: Return a NaN when using operator ( ** ) on Windows version with GPU

[GitBox]

astonzhang commented on issue #9572: Return a NaN when using operator ( ** ) on 
Windows version with GPU
URL: 
https://github.com/apache/incubator-mxnet/issues/9572#issuecomment-360946922
 
 
   @eric-haibin-lin Who do you think is the best person to look into this? 
Thanks.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha opened a new issue #9588: metric should have TP FP TN FN Precision Recall F1 for both macro and micro versions

[GitBox]

szha opened a new issue #9588: metric should have TP FP TN FN Precision Recall 
F1 for both macro and micro versions
URL: https://github.com/apache/incubator-mxnet/issues/9588
 
 
   Related to #9587. The current logic only supports what's called "macro F1" 
logic where we take the mean of F1 scores for mini-batches. "micro F1" should 
also be supported by keeping the tp, fp, fn counts across batches.
   
   Precision and Recall should be added separately.
   5. TopK version of P R F1 should also be supported.
   
   
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/metric.py#L491-L493


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on issue #9572: Return a NaN when using operator ( ** ) on Windows version with GPU

[GitBox]

szha commented on issue #9572: Return a NaN when using operator ( ** ) on 
Windows version with GPU
URL: 
https://github.com/apache/incubator-mxnet/issues/9572#issuecomment-360947107
 
 
   @yajiedesign @zhreshold would either of you take a look? Thanks.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] astonzhang commented on issue #9572: Return a NaN when using operator ( ** ) on Windows version with GPU

[GitBox]

astonzhang commented on issue #9572: Return a NaN when using operator ( ** ) on 
Windows version with GPU
URL: 
https://github.com/apache/incubator-mxnet/issues/9572#issuecomment-360946922
 
 
   @eric-haibin-lin Who do you think is the best person to look into this? 
Thanks.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha opened a new issue #9587: metric.F1 doc needs clarification

[GitBox]

szha opened a new issue #9587: metric.F1 doc needs clarification
URL: https://github.com/apache/incubator-mxnet/issues/9587
 
 
   The current logic only supports what's called "macro F1" logic where we take 
the mean of F1 scores for mini-batches, which needs to be clarified in doc.
   
   
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/metric.py#L491-L493


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] yzhliu commented on issue #9579: Reverting namespace change

[GitBox]

yzhliu commented on issue #9579: Reverting namespace change
URL: https://github.com/apache/incubator-mxnet/pull/9579#issuecomment-360946658
 
 
   Thanks @nswamy 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha opened a new issue #9586: mx.metric F1 is using numpy logic

[GitBox]

szha opened a new issue #9586: mx.metric F1 is using numpy logic
URL: https://github.com/apache/incubator-mxnet/issues/9586
 
 
   The metric module has been using numpy logic and is not benefiting from 
existing mxnet operators.
   
   
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/metric.py#L535-L569


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha opened a new issue #9585: TopKAccuracy is using numpy logic

[GitBox]

szha opened a new issue #9585: TopKAccuracy is using numpy logic
URL: https://github.com/apache/incubator-mxnet/issues/9585
 
 
   The metric module has been using numpy logic and is not benefiting from 
existing mxnet operators.
   
   
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/metric.py#L462-L473


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on a change in pull request #9548: updated release version to 1.0.1 on the mainpage and re-arranged news?

[GitBox]

eric-haibin-lin commented on a change in pull request #9548: updated release 
version to 1.0.1 on the mainpage and re-arranged news?
URL: https://github.com/apache/incubator-mxnet/pull/9548#discussion_r164255583
 
 

 ##
 File path: docs/_static/mxnet-theme/index.html
 ##
 @@ -17,27 +17,26 @@
 
   
 
-
 
   
 
   
-Apache MXNet 1.0 Released
-We're excited to announce the release of MXNet 1.0! Check out the 
release notes for latest updates.
-https://github.com/apache/incubator-mxnet/releases/tag/1.0.0;>Learn 
More
+Apache MXNet 1.0.1 Released
+We're excited to announce the release of MXNet 1.0.1! Check out the 
release notes for latest updates.
+https://github.com/apache/incubator-mxnet/releases/tag/1.0.1;>Learn 
More
 
 Review comment:
   @mbaijal are we supposed to have `1.1.0.rc0` in the link or `1.1.0`?


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on a change in pull request #9548: updated release version to 1.0.1 on the mainpage and re-arranged news?

[GitBox]

eric-haibin-lin commented on a change in pull request #9548: updated release 
version to 1.0.1 on the mainpage and re-arranged news?
URL: https://github.com/apache/incubator-mxnet/pull/9548#discussion_r164255462
 
 

 ##
 File path: docs/_static/mxnet-theme/index.html
 ##
 @@ -17,27 +17,26 @@
 
   
 
-
 
   
 
   
-Apache MXNet 1.0 Released
-We're excited to announce the release of MXNet 1.0! Check out the 
release notes for latest updates.
-https://github.com/apache/incubator-mxnet/releases/tag/1.0.0;>Learn 
More
+Apache MXNet 1.0.1 Released
 
 Review comment:
   Please update the website version to 1.1.0 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] Ishitori commented on issue #9582: Misleading calculation of mxnet.metric.Accuracy

[GitBox]

Ishitori commented on issue #9582: Misleading calculation of 
mxnet.metric.Accuracy
URL: 
https://github.com/apache/incubator-mxnet/issues/9582#issuecomment-360945239
 
 
   It shouldn't be a problem under a correct usage, but there is a way to use 
it incorrectly and system doesn't report that. If we could report it, that 
would increase an ease of use. 
   
   Anyway, it is up to you guys to decide if it is worth fixing.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin opened a new pull request #9584: bump up version to 1.1

[GitBox]

eric-haibin-lin opened a new pull request #9584: bump up version to 1.1
URL: https://github.com/apache/incubator-mxnet/pull/9584
 
 
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on issue #9582: Misleading calculation of mxnet.metric.Accuracy

[GitBox]

szha commented on issue #9582: Misleading calculation of mxnet.metric.Accuracy
URL: 
https://github.com/apache/incubator-mxnet/issues/9582#issuecomment-360944566
 
 
   Is there a case where casting would be a problem under proper usage?


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] Ishitori commented on issue #9582: Misleading calculation of mxnet.metric.Accuracy

[GitBox]

Ishitori commented on issue #9582: Misleading calculation of 
mxnet.metric.Accuracy
URL: 
https://github.com/apache/incubator-mxnet/issues/9582#issuecomment-360944095
 
 
   Agree, it is an incorrect usage of a class. Yet, silent casting is 
happening, which makes it hard to spot a problem. Shouldn't we try to make 
mxnet prone to mistakes like that?


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on issue #9582: Misleading calculation of mxnet.metric.Accuracy

[GitBox]

szha commented on issue #9582: Misleading calculation of mxnet.metric.Accuracy
URL: 
https://github.com/apache/incubator-mxnet/issues/9582#issuecomment-360943676
 
 
   The accuracy class is not intended to be used for one-hot labels. I'm 
clarifying that in the api doc in the above PR. In terms of feature, I think it 
would be fairly easy for the user to do argmax on one-hot label and then feed 
that into metric.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha opened a new pull request #9583: use nd for accuracy calculation

[GitBox]

szha opened a new pull request #9583: use nd for accuracy calculation
URL: https://github.com/apache/incubator-mxnet/pull/9583
 
 
   ## Description ##
   (Brief description on what this PR is about)
   
   ## Checklist ##
   ### Essentials ###
   - [x] Passed code style checking (`make lint`)
   - [x] Changes are complete (i.e. I finished coding on this PR)
   - [x] All changes have test coverage:
   - Unit tests are added for small changes to verify correctness (e.g. adding 
a new operator)
   - Nightly tests are added for complicated/long-running ones (e.g. changing 
distributed kvstore)
   - Build tests will be added for build configuration changes (e.g. adding a 
new build option with NCCL)
   - [x] Code is well-documented: 
   - For user-facing API changes, API doc string has been updated. 
   - For new C++ functions in header files, their functionalities and arguments 
are documented. 
   - For new examples, README.md is added to explain the what the example does, 
the source of the dataset, expected performance on test set and reference to 
the original paper if applicable
   - [x] To the my best knowledge, examples are either not affected by this 
change, or have been fixed to be compatible with this change
   
   ### Changes ###
   - [x] use ndarray instead of numpy for `metric.Accuracy` that fixes #9571 
   - [x] Improve accuracy doc that fixes #9582 
   
   ## Comments ##
   - Backward compatible.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] gpu commented on issue #9572: Return a NaN when using operator ( ** ) on Windows version with GPU

[GitBox]

gpu commented on issue #9572: Return a NaN when using operator ( ** ) on 
Windows version with GPU
URL: 
https://github.com/apache/incubator-mxnet/issues/9572#issuecomment-360942737
 
 
   Thanks for the notifications. But I'm afraid I cannot help you here :o) 
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on issue #8444: /include/mxnet-cpp/operator.hpp:157:31: error: expected ';' at end of declaration std::vector outputs{output};

[GitBox]

szha commented on issue #8444: /include/mxnet-cpp/operator.hpp:157:31: error:   
 expected ';' at end of declaration   std::vector outputs{output};
URL: 
https://github.com/apache/incubator-mxnet/issues/8444#issuecomment-360941899
 
 
   @apache/mxnet-committers: This issue has been inactive for the past 90 days. 
It has no label and needs triage.
   
   For general "how-to" questions, our [user forum](https://discuss.mxnet.io/) 
(and [Chinese version](https://discuss.gluon.ai/)) is a good place to get help.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on issue #8438: HybridBlocks don't work with constant NDArray initialized Parameters

[GitBox]

szha commented on issue #8438: HybridBlocks don't work with constant NDArray 
initialized Parameters
URL: 
https://github.com/apache/incubator-mxnet/issues/8438#issuecomment-360941902
 
 
   @apache/mxnet-committers: This issue has been inactive for the past 90 days. 
It has no label and needs triage.
   
   For general "how-to" questions, our [user forum](https://discuss.mxnet.io/) 
(and [Chinese version](https://discuss.gluon.ai/)) is a good place to get help.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on issue #8448: How can I get symbol shape of a layer as input of another layer in run time? Not with ?infer_shape?

[GitBox]

szha commented on issue #8448: How can I get symbol shape of a layer as input 
of another layer in run time? Not with ?infer_shape?
URL: 
https://github.com/apache/incubator-mxnet/issues/8448#issuecomment-360941895
 
 
   @apache/mxnet-committers: This issue has been inactive for the past 90 days. 
It has no label and needs triage.
   
   For general "how-to" questions, our [user forum](https://discuss.mxnet.io/) 
(and [Chinese version](https://discuss.gluon.ai/)) is a good place to get help.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] pracheer commented on issue #8749: [FIX ATTACHED] Not working example on Iterators tutorial - SimpleIter

[GitBox]

pracheer commented on issue #8749: [FIX ATTACHED] Not working example on 
Iterators tutorial - SimpleIter
URL: 
https://github.com/apache/incubator-mxnet/issues/8749#issuecomment-360941877
 
 
   This issues has been already addressed by 
https://github.com/apache/incubator-mxnet/pull/9460.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] Ishitori opened a new issue #9582: Misleading calculation of mxnet.metric.Accuracy

[GitBox]

Ishitori opened a new issue #9582: Misleading calculation of 
mxnet.metric.Accuracy
URL: https://github.com/apache/incubator-mxnet/issues/9582
 
 
   ## Description
   When passes "preds" and "labels" arguments of the same shape to 
mxnet.metric.Accuracy.update() method, they get silently converted to ints and 
compared element-wise. This can lead to a weird behavior when "labels" actually 
are in one-hot encoded format. See the case on StackOverflow: 
https://stackoverflow.com/questions/45798125/why-mxnet-is-reporting-the-incorrect-validation-accuracy/48471241
   
   I recommend add a warning message if the conversion is happening from floats 
to int. Removing the conversion is also possible, but it is a breaking change.
   
   ## Environment info (Required)
   
   ```
   --Python Info--
   Version  : 3.6.2
   Compiler : GCC 4.2.1 Compatible Apple LLVM 8.1.0 (clang-802.0.42)
   Build: ('default', 'Jul 17 2017 16:44:45')
   Arch : ('64bit', '')
   Pip Info---
   Version  : 9.0.1
   Directory: /usr/local/lib/python3.6/site-packages/pip
   --MXNet Info---
   Version  : 0.11.0
   Directory: /usr/local/lib/python3.6/site-packages/mxnet
   Commit Hash   : 53274b4a2b0d73f3fbdb10cfb5f9ed0c8263fda7
   --System Info--
   Platform : Darwin-16.7.0-x86_64-i386-64bit
   system   : Darwin
   node : 
   release  : 16.7.0
   version  : Darwin Kernel Version 16.7.0: Thu Jan 11 22:59:40 PST 2018; 
root:xnu-3789.73.8~1/RELEASE_X86_64
   --Hardware Info--
   machine  : x86_64
   processor: i386
   b'machdep.cpu.extfeatures: SYSCALL XD 1GBPAGE EM64T LAHF LZCNT PREFETCHW 
RDTSCP TSCI'
   b'machdep.cpu.leaf7_features: SMEP ERMS RDWRFSGS TSC_THREAD_OFFSET BMI1 HLE 
AVX2 BMI2 INVPCID RTM SMAP RDSEED ADX IPT SGX FPU_CSDS MPX CLFSOPT'
   b'machdep.cpu.features: FPU VME DE PSE TSC MSR PAE MCE CX8 APIC SEP MTRR PGE 
MCA CMOV PAT PSE36 CLFSH DS ACPI MMX FXSR SSE SSE2 SS HTT TM PBE SSE3 PCLMULQDQ 
DTES64 MON DSCPL VMX SMX EST TM2 SSSE3 FMA CX16 TPR PDCM SSE4.1 SSE4.2 x2APIC 
MOVBE POPCNT AES PCID XSAVE OSXSAVE SEGLIM64 TSCTMR AVX1.0 RDRAND F16C'
   b'machdep.cpu.brand_string: Intel(R) Core(TM) i7-7660U CPU @ 2.50GHz'
   --Network Test--
   Setting timeout: 10
   Timing for MXNet: https://github.com/apache/incubator-mxnet, DNS: 0.0030 
sec, LOAD: 0.6859 sec.
   Timing for Gluon Tutorial(en): http://gluon.mxnet.io, DNS: 0.0009 sec, LOAD: 
0.1113 sec.
   Timing for Gluon Tutorial(cn): https://zh.gluon.ai, DNS: 0.0015 sec, LOAD: 
0.2410 sec.
   Timing for FashionMNIST: 
https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/dataset/fashion-mnist/train-labels-idx1-ubyte.gz,
 DNS: 0.0007 sec, LOAD: 0.2651 sec.
   Timing for PYPI: https://pypi.python.org/pypi/pip, DNS: 0.0008 sec, LOAD: 
0.2126 sec.
   Timing for Conda: https://repo.continuum.io/pkgs/free/, DNS: 0.0006 sec, 
LOAD: 0.1419 sec.
   ```
   
   ## Error Message:
   There is no visible error message
   
   ## Minimum reproducible example
   ```
   import mxnet as mx
   predicts = mx.nd.array([[1.29206967e-09,   3.40120096e-05,   2.23299547e-12, 
  3.98692492e-07,
   1.21151755e-10,   2.59370694e-08,   1.95488334e-02,   1.13474562e-05,
   9.80405331e-01,   3.51648767e-12]])
   labels = mx.nd.array([[0, 0, 0, 0, 0, 0, 0, 0, 1, 0]])
   acc = mx.metric.Accuracy()
   acc.update(preds=predicts, labels=labels)
   print(acc.get())
   ```
   
   which gives us:
   
   `('accuracy', 0.90002)`
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] anirudh2290 opened a new pull request #9581: Force the dtype of data to be float32 for test_correlation

[GitBox]

anirudh2290 opened a new pull request #9581: Force the dtype of data to be 
float32 for test_correlation
URL: https://github.com/apache/incubator-mxnet/pull/9581
 
 
   ## Description ##
   Force the dtype of img1 and img2 to be float32. This should fix the 
intermittent breaking of test: test_correlation.
   
   @eric-haibin-lin @marcoabreu 
   
   ## Checklist ##
   ### Essentials ###
   - [x] Passed code style checking (`make lint`)
   - [x] Changes are complete (i.e. I finished coding on this PR)
   - [x] All changes have test coverage:
   - Unit tests are added for small changes to verify correctness (e.g. adding 
a new operator)
   - Nightly tests are added for complicated/long-running ones (e.g. changing 
distributed kvstore)
   - Build tests will be added for build configuration changes (e.g. adding a 
new build option with NCCL)
   - [x] Code is well-documented: 
   - For user-facing API changes, API doc string has been updated. 
   - For new C++ functions in header files, their functionalities and arguments 
are documented. 
   - For new examples, README.md is added to explain the what the example does, 
the source of the dataset, expected performance on test set and reference to 
the original paper if applicable
   - [x] To the my best knowledge, examples are either not affected by this 
change, or have been fixed to be compatible with this change


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on a change in pull request #9575: replaced how_to with faq

[GitBox]

szha commented on a change in pull request #9575: replaced how_to with faq
URL: https://github.com/apache/incubator-mxnet/pull/9575#discussion_r164245730
 
 

 ##
 File path: docs/build_version_doc/build_doc.sh
 ##
 @@ -70,6 +70,14 @@ then
 cp -a "$local_build/." "$web_folder"
 fi
 
+# Date Added: 01/26/2018
+# Feature:If a change is requested on the asf-site (stable/root/live 
version), do the following:
+# 1. Checkout the latest branch (eg: 1.0.0)
+# 2. Update
+# 
+
+if 
 
 Review comment:
   @thinksanky ping


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zhanghang1989 closed issue #9580: BatchNorm Not Backward Correctly in Evaluation Mode

[GitBox]

zhanghang1989 closed issue #9580: BatchNorm Not Backward Correctly in 
Evaluation Mode
URL: https://github.com/apache/incubator-mxnet/issues/9580
 
 
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zhanghang1989 commented on issue #9580: BatchNorm Not Backward Correctly in Evaluation Mode

[GitBox]

zhanghang1989 commented on issue #9580: BatchNorm Not Backward Correctly in 
Evaluation Mode
URL: 
https://github.com/apache/incubator-mxnet/issues/9580#issuecomment-360928440
 
 
   Solved. Should use ``loss1.backward(train_mode=False)``


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zhreshold commented on a change in pull request #8918: Added in Large-Batch SGD with a warmup, and a LARS startegy. Also add?

[GitBox]

zhreshold commented on a change in pull request #8918: Added in Large-Batch SGD 
with a warmup, and a LARS startegy. Also add?
URL: https://github.com/apache/incubator-mxnet/pull/8918#discussion_r164242034
 
 

 ##
 File path: python/mxnet/lr_scheduler.py
 ##
 @@ -136,3 +136,42 @@ def __call__(self, num_update):
 else:
 return self.base_lr
 return self.base_lr
+
+class PolyScheduler(LRScheduler):
+""" Reduce the learning rate by given a list of steps.
+
+Calculate the new learning rate by::
+
+   base_lr * (1-nup/max_nup)^pwr
+   if nup < max_nup, 0 otherwise.
+
+Parameters
+--
+   num_update: current number of updates
+   max_update: maximum number of updates before the decay reaches 0.
+   base_lr:base learning rate
+   pwr:   power of the decay term as a funtion of the current number of 
updates.
+
+"""
+
+def __init__(self, num_update, max_update, base_lr=0.01, pwr=2):
+super(PolyScheduler, self).__init__(base_lr)
+assert isinstance(max_update, int)
+if max_update < 1:
+raise ValueError("maximum number of updates must be strictly 
positive")
+self.base_lr_orig = self.base_lr
+self.max_update = max_update
+self.power = pwr
+self.count = num_update
+if num_update <= max_update:
 
 Review comment:
   This this duplicate with line 173. I understand it is for resume training, 
but that should be handled in `__call__`, see the example in 
MultiFactorScheduler. Therefore, num_update is not necessary in `__init__`


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] aaronmarkham commented on issue #9535: Tutorials index page update

[GitBox]

aaronmarkham commented on issue #9535: Tutorials index page update
URL: https://github.com/apache/incubator-mxnet/pull/9535#issuecomment-360927660
 
 
   Based on feedback, I simplified the selector to reflect the current 
arrangement and ordering of the 22 Gluon/MXNet (v1.0) tutorials. 
   
   One of these v1.0 tutorials is commented out because it is broken (#9532).
   
   Now we do not have topical categories. The seven topics such as human 
language and image recognition are commented out for now. We can return to 
discussing their placement, the tutorials within, ordering, etc in the next 
release.
   
   I retained the addition of the 13 tutorials for C++, Scala, and R. 
   
   My first draft had 46 Python tutorials arrange into topics, but since we 
need alignment on what to add I've reduced this to 33 tutorials by removing any 
Gluon tutorials from The Straight Dope that were not in v1.0.
   
   Overall this still gives us an increase of over 20 tutorials that were fixed 
during the usability sprint. Most of these are now in the "Applications" 
grouping. They are alphabetized. 
   
   Note: It isn't possible to use category headings in markdown to break these 
apart, since they'll always appear. This is from the selector code problems... 
which is why I had forced tutorials into categories that we generally related 
and not exactly, technically a perfect match.
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zhanghang1989 opened a new issue #9580: BatchNorm Not Backward Correctly in Evaluation Mode

[GitBox]

zhanghang1989 opened a new issue #9580: BatchNorm Not Backward Correctly in 
Evaluation Mode
URL: https://github.com/apache/incubator-mxnet/issues/9580
 
 
   ## Description
   
   I found the evaluation mode backward is already supported in the source code 
https://github.com/apache/incubator-mxnet/blob/master/src/operator/nn/batch_norm.cu#L368-L375,
 but it is not called due to the wrong context given during the execution.
   
   ## Steps to reproduce
   add  ``LOG(INFO) << "SetupFlags: ctx.is_train : " << ctx.is_train;`` to 
SetupFlags function 
https://github.com/apache/incubator-mxnet/blob/master/src/operator/nn/batch_norm.cu#L578
   
   compile MXNet with ``USE_CUDNN=0`` to make sure not calling cudnn
   
   ## Minimum reproducible example
   ```python
   x = 
(mx.nd.random.uniform(0,1,shape=(B,C,H,W))).square().as_in_context(mx.gpu(0))
   layer1 = nn.BatchNorm(in_channels = C)
   layer1.initialize(ctx = mx.gpu(0))
   with autograd.record(train_mode=False):
   y1 = layer1(x)
   loss1 = y1.sum()
   loss1.backward()
   ```
   
   ## Output from the terminal
   [22:51:47] src/operator/nn/batch_norm.cu:579: SetupFlags: ctx.is_train : 0
   [22:51:47] src/operator/nn/batch_norm.cu:579: SetupFlags: ctx.is_train : 1
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[incubator-mxnet] branch master updated: fix vision datasets doc (#9576)

[haibin]

This is an automated email from the ASF dual-hosted git repository.

haibin pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git


The following commit(s) were added to refs/heads/master by this push:
 new 7fb8154  fix vision datasets doc (#9576)
7fb8154 is described below

commit 7fb8154eefbc631830515ea43a22af48c0eac1c6
Author: Sheng Zha 
AuthorDate: Fri Jan 26 14:41:52 2018 -0800

fix vision datasets doc (#9576)
---
 docs/api/python/gluon/data.md | 7 ++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/docs/api/python/gluon/data.md b/docs/api/python/gluon/data.md
index 0b5f959..28433c0 100644
--- a/docs/api/python/gluon/data.md
+++ b/docs/api/python/gluon/data.md
@@ -51,8 +51,10 @@ In the rest of this document, we list routines provided by 
the `gluon.data` pack
 
 ### Vision
 
+ Vision Datasets
+
 ```eval_rst
-.. currentmodule:: mxnet.gluon.data.vision
+.. currentmodule:: mxnet.gluon.data.vision.datasets
 ```
 
 ```eval_rst
@@ -80,6 +82,9 @@ In the rest of this document, we list routines provided by 
the `gluon.data` pack
 .. automodule:: mxnet.gluon.data.vision
 :members:
 
+.. automodule:: mxnet.gluon.data.vision.datasets
+:members:
+
 ```
 
 auto_index("api-reference");

-- 
To stop receiving notification emails like this one, please contact
hai...@apache.org.

[GitHub] eric-haibin-lin closed pull request #9576: fix vision datasets doc

[GitBox]

eric-haibin-lin closed pull request #9576: fix vision datasets doc
URL: https://github.com/apache/incubator-mxnet/pull/9576
 
 
   

This is a PR merged from a forked repository.
As GitHub hides the original diff on merge, it is displayed below for
the sake of provenance:

As this is a foreign pull request (from a fork), the diff is supplied
below (as it won't show otherwise due to GitHub magic):

diff --git a/docs/api/python/gluon/data.md b/docs/api/python/gluon/data.md
index 0b5f959e32..28433c0ae3 100644
--- a/docs/api/python/gluon/data.md
+++ b/docs/api/python/gluon/data.md
@@ -51,8 +51,10 @@ In the rest of this document, we list routines provided by 
the `gluon.data` pack
 
 ### Vision
 
+ Vision Datasets
+
 ```eval_rst
-.. currentmodule:: mxnet.gluon.data.vision
+.. currentmodule:: mxnet.gluon.data.vision.datasets
 ```
 
 ```eval_rst
@@ -80,6 +82,9 @@ In the rest of this document, we list routines provided by 
the `gluon.data` pack
 .. automodule:: mxnet.gluon.data.vision
 :members:
 
+.. automodule:: mxnet.gluon.data.vision.datasets
+:members:
+
 ```
 
 auto_index("api-reference");


 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] marcoabreu commented on issue #9579: Reverting namespace change

[GitBox]

marcoabreu commented on issue #9579: Reverting namespace change
URL: https://github.com/apache/incubator-mxnet/pull/9579#issuecomment-360923184
 
 
   @srochel 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] thomelane commented on issue #7943: Request: Visualization for Gluon neural networks

[GitBox]

thomelane commented on issue #7943: Request: Visualization for Gluon neural 
networks
URL: 
https://github.com/apache/incubator-mxnet/issues/7943#issuecomment-360896662
 
 
   You're able to plot if you've got a HybridBlock, since you can convert to a 
Symbol by hybridizing and passing an input Symbol. A more direct method might 
be nice though.
   
   ```
   import mxnet as mx
   
   net = mx.gluon.nn.HybridSequential()
   with net.name_scope():
   net.add(mx.gluon.nn.Dense(256, activation="relu"))
   net.add(mx.gluon.nn.Dense(256, activation="relu"))
   net.add(mx.gluon.nn.Dense(10))
   
   # Option 1
   print(net)
   
   # Option 2 (net must be HybridSequential, if want to plot whole graph)
   net.hybridize()
   net.collect_params().initialize()
   x = mx.sym.var('data')
   sym = net(x)
   mx.viz.plot_network(sym)
   ```


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] thomelane commented on issue #7943: Request: Visualization for Gluon neural networks

[GitBox]

thomelane commented on issue #7943: Request: Visualization for Gluon neural 
networks
URL: 
https://github.com/apache/incubator-mxnet/issues/7943#issuecomment-360896662
 
 
   Until this is added, you might be able to work with `_cached_graph` property 
of HybridBlocks.
   
   ```
   import mxnet as mx
   
   net = mx.gluon.nn.HybridSequential()
   with net.name_scope():
   net.add(mx.gluon.nn.Dense(256, activation="relu"))
   net.add(mx.gluon.nn.Dense(256, activation="relu"))
   net.add(mx.gluon.nn.Dense(10))
   
   # Option 1
   print(net)
   
   # Option 2 (net must be HybridSequential, if want to plot whole graph)
   net.hybridize()
   net.collect_params().initialize()
   x = mx.sym.var('data')
   sym = net(x)
   mx.viz.plot_network(sym)
   ```


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] nswamy opened a new pull request #9579: Reverting namespace change

[GitBox]

nswamy opened a new pull request #9579: Reverting namespace change
URL: https://github.com/apache/incubator-mxnet/pull/9579
 
 
   Reverting namespace change since it breaks backward compatibility and needs 
a major version bump up.
   
   Revert "Refactor scala code/tests/examples/docs to use org.apache.mxnet 
(#9324)"
   
   This reverts commit deb25bca9803cac338faedb2eb14dfda22813b75.
   
   ## Description ##
   (Brief description on what this PR is about)
   
   ## Checklist ##
   ### Essentials ###
   - [ ] Passed code style checking (`make lint`)
   - [ ] Changes are complete (i.e. I finished coding on this PR)
   - [ ] All changes have test coverage:
   - Unit tests are added for small changes to verify correctness (e.g. adding 
a new operator)
   - Nightly tests are added for complicated/long-running ones (e.g. changing 
distributed kvstore)
   - Build tests will be added for build configuration changes (e.g. adding a 
new build option with NCCL)
   - [ ] Code is well-documented: 
   - For user-facing API changes, API doc string has been updated. 
   - For new C++ functions in header files, their functionalities and arguments 
are documented. 
   - For new examples, README.md is added to explain the what the example does, 
the source of the dataset, expected performance on test set and reference to 
the original paper if applicable
   - [ ] To the my best knowledge, examples are either not affected by this 
change, or have been fixed to be compatible with this change
   
   ### Changes ###
   - [ ] Feature1, tests, (and when applicable, API doc)
   - [ ] Feature2, tests, (and when applicable, API doc)
   
   ## Comments ##
   - If this change is a backward incompatible change, why must this change be 
made.
   - Interesting edge cases to note here
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zheng-da commented on a change in pull request #8302: Refactor operators & MKLDNN

[GitBox]

zheng-da commented on a change in pull request #8302: Refactor operators & 
MKLDNN
URL: https://github.com/apache/incubator-mxnet/pull/8302#discussion_r164232807
 
 

 ##
 File path: src/kvstore/kvstore_dist.h
 ##
 @@ -32,11 +32,6 @@
 #include "mxnet/engine.h"
 #include "ps/ps.h"
 #include "./kvstore_dist_server.h"
-#if MKL_EXPERIMENTAL == 1
 
 Review comment:
   I tried `../../tools/launch.py --launcher local -n 3 python train_cifar10.py 
--kv-store dist_sync`. I let it run for a long time. It seems to converge well.


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] anirudh2290 commented on a change in pull request #9570: Add checksum for file download in test_consistency

[GitBox]

anirudh2290 commented on a change in pull request #9570: Add checksum for file 
download in test_consistency
URL: https://github.com/apache/incubator-mxnet/pull/9570#discussion_r164232269
 
 

 ##
 File path: tests/python/gpu/test_forward.py
 ##
 @@ -41,8 +42,19 @@ def _dump_images(shape):
 np.save('data/test_images_%d_%d.npy'%shape, imgs)
 
 def _get_data(shape):
-download("http://data.mxnet.io/data/test_images_%d_%d.npy; % (shape), 
dirname='data')
-download("http://data.mxnet.io/data/inception-v3-dump.npz;, dirname="data")
+hash_test_img = "355e15800642286e7fe607d87c38aeeab085b0cc"
+hash_inception_v3 = "91807dfdbd336eb3b265dd62c2408882462752b9"
+fname = utils.download("http://data.mxnet.io/data/test_images_%d_%d.npy; % 
(shape),
+   path="data/test_images_%d_%d.npy" % (shape),
+   sha1_hash=hash_test_img)
+if not utils.check_sha1(fname, hash_test_img):
+raise RuntimeError("File %s not downloaded completely" % 
("test_images_%d_%d.npy"%(shape)))
+
+fname = utils.download("http://data.mxnet.io/data/inception-v3-dump.npz;,
+   path='data/inception-v3-dump.npz',
+   sha1_hash=hash_inception_v3)
+if not utils.check_sha1(fname, hash_inception_v3):
+raise RuntimeError("File %s not downloaded completely" % 
("inception-v3-dump.npz"))
 
 Review comment:
   Ah. Thanks for pointing! Fixed in #9578 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] anirudh2290 opened a new pull request #9578: Remove non-necessary RuntimeError

[GitBox]

anirudh2290 opened a new pull request #9578: Remove non-necessary RuntimeError
URL: https://github.com/apache/incubator-mxnet/pull/9578
 
 
   ## Description ##
   utils.download already raises an Exception if the checksum doesn't match. 
Not needed to do it twice. 
   
   @szha @eric-haibin-lin 
   
   ## Checklist ##
   ### Essentials ###
   - [ ] Passed code style checking (`make lint`)
   - [ ] Changes are complete (i.e. I finished coding on this PR)
   - [ ] All changes have test coverage:
   - Unit tests are added for small changes to verify correctness (e.g. adding 
a new operator)
   - Nightly tests are added for complicated/long-running ones (e.g. changing 
distributed kvstore)
   - Build tests will be added for build configuration changes (e.g. adding a 
new build option with NCCL)
   - [ ] Code is well-documented: 
   - For user-facing API changes, API doc string has been updated. 
   - For new C++ functions in header files, their functionalities and arguments 
are documented. 
   - For new examples, README.md is added to explain the what the example does, 
the source of the dataset, expected performance on test set and reference to 
the original paper if applicable
   - [ ] To the my best knowledge, examples are either not affected by this 
change, or have been fixed to be compatible with this change
   
   ### Changes ###
   - [ ] Feature1, tests, (and when applicable, API doc)
   - [ ] Feature2, tests, (and when applicable, API doc)
   
   ## Comments ##
   - If this change is a backward incompatible change, why must this change be 
made.
   - Interesting edge cases to note here
   


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on issue #9576: fix vision datasets doc

[GitBox]

szha commented on issue #9576: fix vision datasets doc
URL: https://github.com/apache/incubator-mxnet/pull/9576#issuecomment-360912904
 
 
   Before: 
https://mxnet.incubator.apache.org/versions/master/api/python/gluon/data.html#vision
   After: 
http://mxnet-doc.s3-accelerate.dualstack.amazonaws.com/api/python/gluon/data.html#vision


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin opened a new pull request #9577: Disable flaky test: test_correlation

[GitBox]

eric-haibin-lin opened a new pull request #9577: Disable flaky test: 
test_correlation
URL: https://github.com/apache/incubator-mxnet/pull/9577
 
 
   #9553 
   @marcoabreu @anirudh2290 


This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services