From ace10ab15d30d6d567d9f2e960fba24e587c7611 Mon Sep 17 00:00:00 2001
From: Andrew DalPino <me@andrewdalpino.com>
Date: Thu, 6 Aug 2020 18:19:23 -0500
Subject: [PATCH] Update to Rubix ML 0.1.0

---
 README.md     | 28 ++++++++++++++++++++++++++--
 composer.json |  3 +--
 predict.php   |  8 +++-----
 train.php     | 11 +++++------
 4 files changed, 35 insertions(+), 15 deletions(-)

diff --git a/README.md b/README.md
index 8c90949..4646ea9 100644
--- a/README.md
+++ b/README.md
@@ -99,7 +99,7 @@ Since Gradient Boost implements the [Verbose](https://docs.rubixml.com/en/latest
 ```php
 use Rubix\ML\Other\Loggers\Screen;
 
-$estimator->setLogger(new Screen('housing'));
+$estimator->setLogger(new Screen());
 ```
 
 ### Training
@@ -110,7 +110,7 @@ $estimator->train($dataset);
 ```
 
 ### Validation Score and Loss
-During training, the learner will record the validation score and the training loss at each iteration or *epoch*. The validation score is calculated using the default [RMSE](https://docs.rubixml.com/en/latest/cross-validation/metrics/rmse.html) metric on a hold out portion of the training set. Contrariwise, the training loss is the value of the cost function (in this case the L2 or *quadratic* loss) computed over the training data. We can visualize the training progress by plotting these metrics. To export the scores and losses you can call the additional `scores()` and `steps()` methods on the learner instance.
+During training, the learner will record the validation score and the training loss at each iteration or *epoch*. The validation score is calculated using the default [RMSE](https://docs.rubixml.com/en/latest/cross-validation/metrics/rmse.html) metric on a hold out portion of the training set. Contrariwise, the training loss is the value of the cost function (in this case the L2 or *quadratic* loss) computed over the training data. We can visualize the training progress by plotting these metrics. To output the scores and losses you can call the additional `scores()` and `steps()` methods on the learner instance.
 
 ```php
 $scores = $estimator->scores();
@@ -118,6 +118,19 @@ $scores = $estimator->scores();
 $losses = $estimator->steps();
 ```
 
+Then we can export the data to a CSV file using an [Unlabeled](https://docs.rubixml.com/en/latest/datasets/unlabeled.html) dataset object. The `array_transpose()` method takes a 2-dimensional array and changes the rows to columns and vice versa.
+
+```php
+use Rubix\ML\Unlabeled;
+use function Rubix\ML\array_transpose;
+
+Unlabeled::build(array_transpose([$scores, $losses]))
+    ->toCSV(['scores', 'losses'])
+    ->write('progress.csv');
+
+```
+
+
 Here is an example of what the validation score and training loss look like when plotted. You can plot the values yourself by importing the `progress.csv` file into your favorite plotting software.
 
 ![R Squared Score](https://raw.githubusercontent.com/RubixML/Housing/master/docs/images/validation-score.svg?sanitize=true)
@@ -176,6 +189,17 @@ To obtain the predictions from the model, call the `predict()` method with the d
 $predictions = $estimator->predict($dataset);
 ```
 
+Then we'll use another [Unlabeled](https://docs.rubixml.com/en/latest/datasets/unlabeled.html) dataset to write the IDs and predictions to a CSV file that we'll submit to the competition.
+
+```php
+use Rubix\ML\Datasets\Unlabeled;
+use function Rubix\ML\array_transpose;
+
+Unlabeled::build(array_transpose([$ids, $predictions]))
+    ->toCSV(['Id', 'SalePrice'])
+    ->write('predictions.csv');
+```
+
 Now run the prediction script by calling it from the command line.
 ```sh
 $ php predict.php
diff --git a/composer.json b/composer.json
index 4d3a43b..4081ea1 100644
--- a/composer.json
+++ b/composer.json
@@ -20,8 +20,7 @@
     ],
     "require": {
         "php": ">=7.2",
-        "league/csv": "^9.5",
-        "rubix/ml": "^0.1.0-rc2"
+        "rubix/ml": "^0.1.0"
     },
     "suggest": {
         "ext-tensor": "For faster training and inference"
diff --git a/predict.php b/predict.php
index 70f88b7..29d0e33 100644
--- a/predict.php
+++ b/predict.php
@@ -7,7 +7,6 @@
 use Rubix\ML\Transformers\NumericStringConverter;
 use Rubix\ML\PersistentModel;
 use Rubix\ML\Persisters\Filesystem;
-use League\Csv\Writer;
 
 use function Rubix\ML\array_transpose;
 
@@ -28,9 +27,8 @@
 
 $predictions = $estimator->predict($dataset);
 
-$writer = Writer::createFromPath('predictions.csv', 'w+');
-
-$writer->insertOne(['Id', 'SalePrice']);
-$writer->insertAll(array_transpose([$ids, $predictions]));
+Unlabeled::build(array_transpose([$ids, $predictions]))
+    ->toCSV(['Id', 'SalePrice'])
+    ->write('predictions.csv');
 
 echo 'Predictions saved to predictions.csv' . PHP_EOL;
\ No newline at end of file
diff --git a/train.php b/train.php
index 7bef275..db0589a 100644
--- a/train.php
+++ b/train.php
@@ -12,7 +12,7 @@
 use Rubix\ML\Regressors\RegressionTree;
 use Rubix\ML\Persisters\Filesystem;
 use Rubix\ML\Other\Loggers\Screen;
-use League\Csv\Writer;
+use Rubix\ML\Datasets\Unlabeled;
 
 use function Rubix\ML\array_transpose;
 
@@ -50,7 +50,7 @@
     new Filesystem('housing.model', true)
 );
 
-$estimator->setLogger(new Screen('housing'));
+$estimator->setLogger(new Screen());
 
 echo 'Training ...' . PHP_EOL;
 
@@ -59,10 +59,9 @@
 $scores = $estimator->scores();
 $losses = $estimator->steps();
 
-$writer = Writer::createFromPath('progress.csv', 'w+');
-
-$writer->insertOne(['score', 'loss']);
-$writer->insertAll(array_transpose([$scores, $losses]));
+Unlabeled::build(array_transpose([$scores, $losses]))
+    ->toCSV(['scores', 'losses'])
+    ->write('progress.csv');
 
 echo 'Progress saved to progress.csv' . PHP_EOL;