easy-cs-rec-custommodel 0.8.6__py2.py3-none-any.whl

easy_rec/python/test/emr_run.py

@@ -0,0 +1,119 @@
+# -*- encoding:utf-8 -*-
+# Copyright (c) Alibaba, Inc. and its affiliates.
+
+import argparse
+import logging
+import os
+import shutil
+import sys
+
+import tensorflow as tf
+
+from easy_rec.python.test.odps_command import OdpsCommand
+from easy_rec.python.test.odps_test_prepare import prepare
+from easy_rec.python.test.odps_test_util import OdpsOSSConfig
+from easy_rec.python.test.odps_test_util import delete_oss_path
+from easy_rec.python.test.odps_test_util import get_oss_bucket
+from easy_rec.python.utils import test_utils
+
+logging.basicConfig(
+    level=logging.INFO, format='[%(asctime)s][%(levelname)s] %(message)s')
+
+odps_oss_config = OdpsOSSConfig(script_path='./samples/emr_script')
+
+
+class TestPipelineOnEmr(tf.test.TestCase):
+  """Train eval test on emr."""
+
+  def setUp(self):
+    logging.info('Testing %s.%s' % (type(self).__name__, self._testMethodName))
+    self._test_hdfs_dir = test_utils.get_hdfs_tmp_dir(
+        'hdfs://emr-header-1:9000/user/easy_rec/emr_test')
+    self._success = True
+    logging.info('test hdfs dir: %s' % self._test_hdfs_dir)
+
+  def tearDown(self):
+    if self._success:
+      pass
+    test_utils.clean_up_hdfs(self._test_hdfs_dir)
+
+  def test_deepfm_train_eval_export(self):
+    start = [
+        'deep_fm/create_external_deepfm_table.sql',
+        'deep_fm/create_inner_deepfm_table.sql'
+    ]
+    end = ['deep_fm/drop_table.sql']
+    odps_cmd = OdpsCommand(odps_oss_config)
+    odps_cmd.run_list(start)
+    self._success = test_utils.test_hdfs_train_eval(
+        '%s/configs/deepfm.config' % odps_oss_config.temp_dir,
+        '%s/yaml_config/train.paitf.yaml' % odps_oss_config.temp_dir,
+        self._test_hdfs_dir)
+    self.assertTrue(self._success)
+
+    self._success = test_utils.test_hdfs_eval(
+        '%s/configs/deepfm_eval_pipeline.config' % odps_oss_config.temp_dir,
+        '%s/yaml_config/eval.tf.yaml' % odps_oss_config.temp_dir,
+        self._test_hdfs_dir)
+    self.assertTrue(self._success)
+
+    self._success = test_utils.test_hdfs_export(
+        '%s/configs/deepfm_eval_pipeline.config' % odps_oss_config.temp_dir,
+        '%s/yaml_config/export.tf.yaml' % odps_oss_config.temp_dir,
+        self._test_hdfs_dir)
+    self.assertTrue(self._success)
+
+    odps_cmd.run_list(end)
+
+
+if __name__ == '__main__':
+  parser = argparse.ArgumentParser()
+  parser.add_argument(
+      '--odps_config', type=str, default=None, help='odps config path')
+  parser.add_argument(
+      '--oss_config', type=str, default=None, help='ossutilconfig path')
+  parser.add_argument(
+      '--bucket_name', type=str, default=None, help='test oss bucket name')
+  parser.add_argument('--arn', type=str, default=None, help='oss rolearn')
+  parser.add_argument(
+      '--odpscmd', type=str, default='odpscmd', help='odpscmd path')
+  parser.add_argument(
+      '--algo_project', type=str, default=None, help='algo project name')
+  parser.add_argument(
+      '--algo_res_project',
+      type=str,
+      default=None,
+      help='algo resource project name')
+  parser.add_argument(
+      '--algo_version', type=str, default=None, help='algo version')
+  args, unknown_args = parser.parse_known_args()
+  sys.argv = [sys.argv[0]]
+  for unk_arg in unknown_args:
+    sys.argv.append(unk_arg)
+
+  if args.odps_config:
+    odps_oss_config.load_odps_config(args.odps_config)
+    os.environ['ODPS_CONFIG_FILE_PATH'] = args.odps_config
+  if args.oss_config:
+    odps_oss_config.load_oss_config(args.oss_config)
+  if args.odpscmd:
+    odps_oss_config.odpscmd_path = args.odpscmd
+  if args.algo_project:
+    odps_oss_config.algo_project = args.algo_project
+  if args.algo_res_project:
+    odps_oss_config.algo_res_project = args.algo_res_project
+  if args.algo_version:
+    odps_oss_config.algo_version = args.algo_version
+  if args.arn:
+    odps_oss_config.arn = args.arn
+  if args.bucket_name:
+    odps_oss_config.bucket_name = args.bucket_name
+  print(args)
+  prepare(odps_oss_config)
+  tf.test.main()
+  # delete oss path
+  bucket = get_oss_bucket(odps_oss_config.oss_key, odps_oss_config.oss_secret,
+                          odps_oss_config.endpoint, odps_oss_config.bucket_name)
+  delete_oss_path(bucket, odps_oss_config.exp_dir, odps_oss_config.bucket_name)
+  # delete tmp
+  shutil.rmtree(odps_oss_config.temp_dir)

easy_rec/python/test/eval_metric_test.py

@@ -0,0 +1,107 @@
+# -*- encoding:utf-8 -*-
+# Copyright (c) Alibaba, Inc. and its affiliates.
+from __future__ import division
+
+import logging
+
+import tensorflow as tf
+from absl.testing import parameterized
+
+from easy_rec.python.utils.test_utils import RunAsSubprocess
+
+if tf.__version__ >= '2.0':
+  tf = tf.compat.v1
+
+
+class MetricsTest(tf.test.TestCase, parameterized.TestCase):
+
+  def setUp(self):
+    logging.info('Testing %s.%s' % (type(self).__name__, self._testMethodName))
+
+  @RunAsSubprocess
+  def test_max_f1(self):
+    from easy_rec.python.core.metrics import max_f1
+    labels = tf.constant([1, 0, 0, 1], dtype=tf.int32)
+    probs = tf.constant([0.9, 0.8, 0.7, 0.6], dtype=tf.float32)
+    f1, f1_update_op = max_f1(labels, probs)
+
+    with tf.Session() as sess:
+      sess.run(tf.local_variables_initializer())
+      sess.run(f1_update_op)
+      f1_score = sess.run(f1)
+    self.assertAlmostEqual(f1_score, 2.0 / 3)
+
+  @RunAsSubprocess
+  def test_gauc_all_negative_label(self):
+    from easy_rec.python.core.metrics import gauc
+    labels = tf.constant([0, 0, 0, 0], dtype=tf.int32)
+    probs = tf.constant([0.9, 0.8, 0.7, 0.6], dtype=tf.float32)
+    uids = tf.constant([1, 1, 1, 1], dtype=tf.int32)
+    value_op, update_op = gauc(labels, probs, uids)
+    with tf.Session() as sess:
+      sess.run(update_op)
+      score = sess.run(value_op)
+    self.assertAlmostEqual(score, 0.0)
+
+  @parameterized.named_parameters(
+      [['_reduction_mean', 'mean', 0.5833333],
+       ['_reduction_mean_by_sample_num', 'mean_by_sample_num', 0.5925926],
+       ['_reduction_mean_by_positive_num', 'mean_by_positive_num', 0.6]])
+  @RunAsSubprocess
+  def test_gauc(self, reduction, expected):
+    from easy_rec.python.core.metrics import gauc
+    labels = tf.placeholder(dtype=tf.int32, shape=(None,))
+    probs = tf.placeholder(dtype=tf.float32, shape=(None,))
+    uids = tf.placeholder(dtype=tf.int32, shape=(None,))
+    value_op, update_op = gauc(labels, probs, uids, reduction=reduction)
+    with tf.Session() as sess:
+      sess.run(
+          update_op,
+          feed_dict={
+              labels: [1, 0, 1, 1, 0],
+              probs: [0.9, 0.8, 0.7, 0.6, 0.5],
+              uids: [1, 1, 1, 1, 1]
+          })
+      sess.run(
+          update_op,
+          feed_dict={
+              labels: [1, 0, 0, 1],
+              probs: [0.9, 0.8, 0.7, 0.6],
+              uids: [2, 2, 2, 2]
+          })
+      score = sess.run(value_op)
+    self.assertAlmostEqual(score, expected)
+
+  @parameterized.named_parameters(
+      [['_reduction_mean', 'mean', 0.5833333],
+       ['_reduction_mean_by_sample_num', 'mean_by_sample_num', 0.5925926],
+       ['_reduction_mean_by_positive_num', 'mean_by_positive_num', 0.6]])
+  @RunAsSubprocess
+  def test_session_auc(self, reduction, expected):
+    from easy_rec.python.core.metrics import session_auc
+    labels = tf.placeholder(dtype=tf.int32, shape=(None,))
+    probs = tf.placeholder(dtype=tf.float32, shape=(None,))
+    session_ids = tf.placeholder(dtype=tf.int32, shape=(None,))
+    value_op, update_op = session_auc(
+        labels, probs, session_ids, reduction=reduction)
+    with tf.Session() as sess:
+      sess.run(
+          update_op,
+          feed_dict={
+              labels: [1, 0, 1, 1, 0],
+              probs: [0.9, 0.8, 0.7, 0.6, 0.5],
+              session_ids: [1, 1, 1, 1, 1]
+          })
+      sess.run(
+          update_op,
+          feed_dict={
+              labels: [1, 0, 0, 1],
+              probs: [0.9, 0.8, 0.7, 0.6],
+              session_ids: [2, 2, 2, 2]
+          })
+      score = sess.run(value_op)
+    self.assertAlmostEqual(score, expected)
+
+
+if __name__ == '__main__':
+  tf.test.main()

easy_rec/python/test/excel_convert_test.py

@@ -0,0 +1,64 @@
+# -*- encoding:utf-8 -*-
+# Copyright (c) Alibaba, Inc. and its affiliates.
+
+import logging
+import os
+
+import tensorflow as tf
+
+from easy_rec.python.utils import test_utils
+
+
+class ExcelConvertTest(tf.test.TestCase):
+
+  def setUp(self):
+    logging.info('Testing %s.%s' % (type(self).__name__, self._testMethodName))
+
+  def tearDown(self):
+    test_utils.set_gpu_id(None)
+
+  def test_deepfm_convert(self):
+    test_dir = test_utils.get_tmp_dir()
+    logging.info('test dir: %s' % test_dir)
+    pipeline_config_path = os.path.join(test_dir, 'avazu_deepfm_excel.config')
+    convert_cmd = """
+      python -m easy_rec.python.tools.create_config_from_excel
+        --excel_path samples/excel_config/dwd_avazu_ctr_deepfm.xls
+        --model_type deepfm
+        --output_path  %s
+        --train_input_path data/test/dwd_avazu_ctr_deepmodel_10w.csv
+        --eval_input_path data/test/dwd_avazu_ctr_deepmodel_10w.csv
+    """ % pipeline_config_path
+    proc = test_utils.run_cmd(convert_cmd,
+                              '%s/log_%s.txt' % (test_dir, 'convert'))
+    proc.wait()
+    self.assertTrue(proc.returncode == 0)
+    self.assertTrue(
+        test_utils.test_single_train_eval(
+            pipeline_config_path, test_dir=test_dir))
+    test_utils.clean_up(test_dir)
+
+  def test_multi_tower_convert(self):
+    test_dir = test_utils.get_tmp_dir()
+    logging.info('test dir: %s' % test_dir)
+    pipeline_config_path = os.path.join(test_dir, 'avazu_deepfm_excel.config')
+    convert_cmd = """
+      python -m easy_rec.python.tools.create_config_from_excel
+        --excel_path samples/excel_config/dwd_avazu_ctr_deepfm.xls
+        --model_type multi_tower
+        --output_path  %s
+        --train_input_path data/test/dwd_avazu_ctr_deepmodel_10w.csv
+        --eval_input_path data/test/dwd_avazu_ctr_deepmodel_10w.csv
+    """ % pipeline_config_path
+    proc = test_utils.run_cmd(convert_cmd,
+                              '%s/log_%s.txt' % (test_dir, 'convert'))
+    proc.wait()
+    self.assertTrue(proc.returncode == 0)
+    self.assertTrue(
+        test_utils.test_single_train_eval(
+            pipeline_config_path, test_dir=test_dir))
+    test_utils.clean_up(test_dir)
+
+
+if __name__ == '__main__':
+  tf.test.main()