Merge remote-tracking branch 'upstream/master' into sparse

docs/api/python/autograd.md

@@ -9,6 +9,64 @@
 .. warning:: This package is currently experimental and may change in the near future.
 ```
 
+## Overview
+
+The `autograd` package enables automatic
+differentiation of NDArray operations.
+In machine learning applications,
+`autograd` is often used to calculate the gradients
+of loss functions with respect to parameters.
+
+
+### Record vs Pause
+
+`autograd` records computation history on the fly to calculate gradients later.
+This is only enabled inside a `with autograd.record():` block.
+A `with auto_grad.pause()` block can be used inside a `record()` block
+to temporarily disable recording.
+
+To compute gradient with respect to an `NDArray` `x`, first call `x.attach_grad()`
+to allocate space for the gradient. Then, start a `with autograd.record()` block,
+and do some computation. Finally, call `backward()` on the result:
+
+```python
+>>> x = mx.nd.array([1,2,3,4])
+>>> x.attach_grad()
+>>> with mx.autograd.record():
+...     y = x * x + 1
+>>> print(x.grad)
+[ 2.  4.  6.  8.]
+<NDArray 4 @cpu(0)>
+```
+
+
+## Train mode and Predict Mode
+
+Some operators (Dropout, BatchNorm, etc) behave differently in
+when training and when making predictions.
+This can be controled with `train_mode` and `predict_mode` scope.
+
+By default, MXNet is in `predict_mode`.
+A `with autograd.record()` block by default turns on `train_mode`
+(equivalent to ``with autograd.record(train_mode=True)``).
+To compute a gradient in prediction mode (as when generating adversarial examples),
+call record with `train_mode=False` and then call `backward(train_mode=False)`
+
+Although training usually coincides with recording,
+this isn't always the case.
+To control *training* vs *predict_mode* without changing
+*recording* vs *not recording*,
+Use a `with autograd.train_mode():`
+or `with autograd.predict_mode():` block.
+
+Detailed tutorials are available in Part 1 of
+[the MXNet gluon book](http://gluon.mxnet.io/).
+
+
+
+
+
+
 <script type="text/javascript" src='../../_static/js/auto_module_index.js'></script>
 
 ## Autograd

perl-package/AI-MXNet/Changes

@@ -1,5 +1,8 @@
 Revision history for Perl extension AI::MXNet
 
+1.0102 Sun Aug  6 16:55:08 PDT 2017
+        - bugfixes in Image.pm, updated tests, added PearsonCorrelation metric, added Convolutional RNN modules.
+
 1.0101  Sun Jul  2 17:16:01 PDT 2017
         - reworked CachedOp, two new optimizers, auto module reshape, using strings to index the kvstore.
 

perl-package/AI-MXNet/MANIFEST

@@ -10,7 +10,6 @@ examples/cudnn_lstm_bucketing.pl
 Makefile.PL
 Changes
 META.json
-t/test_autograd.t
 t/test_recordio.t
 t/test_random.t
 t/test_init.t

perl-package/AI-MXNet/META.json

@@ -30,7 +30,7 @@
       },
       "runtime" : {
          "requires" : {
-            "AI::MXNetCAPI" : "1.0101",
+            "AI::MXNetCAPI" : "1.0102",
             "AI::NNVMCAPI" : "1.01",
             "Function::Parameters" : "1.0705",
             "GraphViz" : "2.14",
@@ -43,5 +43,5 @@
       }
    },
    "release_status" : "stable",
-   "version" : "1.0101"
+   "version" : "1.0102"
 }

perl-package/AI-MXNet/META.yml

@@ -17,10 +17,10 @@ no_index:
     - t
     - inc
 requires:
-  AI::MXNetCAPI: '1.0101'
+  AI::MXNetCAPI: '1.0102'
   AI::NNVMCAPI: '1.01'
   Function::Parameters: '1.0705'
   GraphViz: '2.14'
   Mouse: v2.1.0
   PDL: '2.007'
-version: '1.0101'
+version: '1.0102'

perl-package/AI-MXNet/Makefile.PL

@@ -19,7 +19,7 @@ my %WriteMakefileArgs = (
   "LICENSE" => "apache_2_0",
   "NAME" => "AI::MXNet",
   "PREREQ_PM" => {
-    "AI::MXNetCAPI" => "1.0101",
+    "AI::MXNetCAPI" => "1.0102",
     "AI::NNVMCAPI" => "1.01",
     "Function::Parameters" => "1.0705",
     "Mouse" => "v2.1.0",
@@ -35,7 +35,7 @@ my %WriteMakefileArgs = (
 
 
 my %FallbackPrereqs = (
-  "AI::MXNetCAPI" => "1.0101",
+  "AI::MXNetCAPI" => "1.0102",
   "AI::NNVMCAPI" => "1.01",
   "Function::Parameters" => "1.0705",
   "Mouse" => "v2.1.0",

perl-package/AI-MXNet/README

@@ -1,5 +1,5 @@
 This archive contains the distribution AI-MXNet,
-version 1.0101:
+version 1.0102:
 
   Perl interface to MXNet machine learning library
 

perl-package/AI-MXNet/lib/AI/MXNet.pm

@@ -46,7 +46,7 @@ use AI::MXNet::Image;
 use AI::MXNet::Contrib;
 use AI::MXNet::Contrib::AutoGrad;
 use AI::MXNet::CachedOp;
-our $VERSION = '1.0101';
+our $VERSION = '1.0102';
 
 sub import
 {

perl-package/AI-MXNet/lib/AI/MXNet/Base.pm

@@ -20,7 +20,7 @@ use strict;
 use warnings;
 use PDL;
 use PDL::Types qw();
-use AI::MXNetCAPI 1.0101;
+use AI::MXNetCAPI 1.0102;
 use AI::NNVMCAPI 1.01;
 use AI::MXNet::Types;
 use Time::HiRes;

perl-package/AI-MXNet/lib/AI/MXNet/Image.pm

@@ -764,7 +764,7 @@ sub BUILD
         {
             chomp($line);
             my @line = split(/\t/, $line);
-            my $label = AI::MXNet::NDArray->array([@line[1..@line-1]]);
+            my $label = AI::MXNet::NDArray->array([@line[1..@line-2]]);
             my $key   = $line[0];
             $imglist{$key} = [$label, $line[-1]];
             push @imgkeys, $key;
@@ -838,6 +838,10 @@ sub BUILD
     {
         $self->aug_list(AI::MXNet::Image->CreateAugmenter(data_shape => $self->data_shape, %{ $self->kwargs//{} }));
     }
+    else
+    {
+        $self->aug_list([]);
+    }
     $self->cur(0);
     $self->reset();
 }
@@ -877,7 +881,7 @@ method next_sample()
         }
         else
         {
-            my ($label, $fname) = $self->imglist->{$idx};
+            my ($label, $fname) = @{ $self->imglist->{$idx} };
             if(not defined $self->imgrec)
             {
                 open(F, $self->path_root . "/$fname") or confess("can't open $fname $!");

perl-package/AI-MXNet/lib/AI/MXNet/Metric.pm

@@ -510,6 +510,55 @@ method update(ArrayRef[AI::MXNet::NDArray] $labels, ArrayRef[AI::MXNet::NDArray]
     }, $labels, $preds);
 }
 
+package AI::MXNet::PearsonCorrelation;
+use Mouse;
+use AI::MXNet::Base;
+extends 'AI::MXNet::EvalMetric';
+has '+name'   => (default => 'pearson-correlation');
+
+=head1 NAME
+
+    AI::MXNet::PearsonCorrelation
+=cut
+
+=head1 DESCRIPTION
+
+    Computes Pearson correlation.
+
+    Parameters
+    ----------
+    name : str
+        Name of this metric instance for display.
+
+    Examples
+    --------
+    >>> $predicts = [mx->nd->array([[0.3, 0.7], [0, 1.], [0.4, 0.6]])]
+    >>> $labels   = [mx->nd->array([[1, 0], [0, 1], [0, 1]])]
+    >>> $pr = mx->metric->PearsonCorrelation()
+    >>> $pr->update($labels, $predicts)
+    >>> print pr->get()
+    ('pearson-correlation', '0.421637061887229')
+=cut
+
+method update(ArrayRef[AI::MXNet::NDArray] $labels, ArrayRef[AI::MXNet::NDArray] $preds)
+{
+    AI::MXNet::Metric::check_label_shapes($labels, $preds);
+    zip(sub {
+        my ($label, $pred) = @_;
+        AI::MXNet::Metric::check_label_shapes($label, $pred);
+        $label = $label->aspdl->flat;
+        $pred  = $pred->aspdl->flat;
+        my ($label_mean, $label_stdv) = ($label->stats)[0, 6];
+        my ($pred_mean, $pred_stdv) = ($pred->stats)[0, 6];
+        $self->sum_metric(
+            $self->sum_metric
+                +
+            ((($label-$label_mean)*($pred-$pred_mean))->sum/$label->nelem)/(($label_stdv*$pred_stdv)->at(0))
+        );
+        $self->num_inst($self->num_inst + 1);
+    }, $labels, $preds);
+}
+
 =head1 DESCRIPTION
 
     Custom evaluation metric that takes a sub ref.
@@ -574,6 +623,7 @@ my %metrics = qw/
     top_k_accuracy AI::MXNet::TopKAccuracy
     Perplexity     AI::MXNet::Perplexity
     perplexity     AI::MXNet::Perplexity
+    pearsonr       AI::MXNet::PearsonCorrelation
 /;
 
 method create(Metric|ArrayRef[Metric] $metric, %kwargs)

perl-package/AI-MXNet/lib/AI/MXNet/Module.pm

@@ -796,14 +796,6 @@ method forward(
 )
 {
     assert($self->binded and $self->params_initialized);
-    # If starting to do the inference, force rebind the module.
-    if($self->label_shapes and not $data_batch->label)
-    {
-        confess(
-            "If you are trying to do inference, rebind module ".
-            "with 'force_rebind=True' and 'for_training=False'"
-        );
-    }
 
     my @curr_data_shapes = map { $_->shape } @{ $self->data_shapes };
     my @new_data_shapes  = map { $_->shape } @{ $data_batch->data };

perl-package/AI-MXNet/lib/AI/MXNet/RNN.pm

@@ -166,6 +166,9 @@ method SequentialRNNCell(@args)  { AI::MXNet::RNN::SequentialCell->new(@args) }
 method BidirectionalCell(@args)  { AI::MXNet::RNN::BidirectionalCell->new(@args) }
 method DropoutCell(@args)        { AI::MXNet::RNN::DropoutCell->new(@args) }
 method ZoneoutCell(@args)        { AI::MXNet::RNN::ZoneoutCell->new(@args) }
+method ConvRNNCell(@args)        { AI::MXNet::RNN::ConvCell->new(@args) }
+method ConvLSTMCell(@args)       { AI::MXNet::RNN::ConvLSTMCell->new(@args) }
+method ConvGRUCell(@args)        { AI::MXNet::RNN::ConvGRUCell->new(@args) }
 method ResidualCell(@args)       { AI::MXNet::RNN::ResidualCell->new(@args) }
 method encode_sentences(@args)   { AI::MXNet::RNN::IO->encode_sentences(@args) }
 method BucketSentenceIter(@args)