From f342a4f6b65116f69e3be99eea561a95e4b18c4a Mon Sep 17 00:00:00 2001
From: dmt <>
Date: Thu, 3 Oct 2019 19:06:53 +0200
Subject: [PATCH] Identify blocks that satisfy the criteria for being a
learning block.
---
cml/domain/data_source.py | 19 +++++++++++++++++++
1 file changed, 19 insertions(+)
diff --git a/cml/domain/data_source.py b/cml/domain/data_source.py
index c438572..d9e6242 100644
--- a/cml/domain/data_source.py
+++ b/cml/domain/data_source.py
@@ -57,3 +57,22 @@ class Preprocessor:
class LearnblockIdentifier:
def __init__(self, settings):
self.settings = settings
+ self.column_pairs = (("T", "Z"), ("T", "Sigma"), ("Sigma", "Z"))
+
+ def identify(self, block):
+ for pair in self.column_pairs:
+ for possible_learnblock in self._identify_relatives(block, *pair):
+ if self._is_learn_block(possible_learnblock.length):
+ yield possible_learnblock
+
+ def _is_learn_block(self, block_length):
+ return block_length > self.settings.learn_block_minimum
+
+ def _identify_relatives(self, block, *args):
+ # TODO (dmt): Implement density and kmeans!
+ already_seen = set()
+ for value_pair in block.get_duplicated_pairs(args[0], args[1]):
+ if value_pair not in already_seen:
+ already_seen.add(value_pair)
+ kw = {args[0]: value_pair[0], args[1]: value_pair[1]}
+ yield block.get_values(**kw)
--
GitLab