diff --git a/.cache/cache.db b/.cache/cache.db
deleted file mode 100644
index d8d3e492c9ea34b5ada4eae9053518cb5534e344..0000000000000000000000000000000000000000
Binary files a/.cache/cache.db and /dev/null differ
diff --git a/data/cache.dat b/data/cache.dat
new file mode 100644
index 0000000000000000000000000000000000000000..317f612fb8476a15fba7fa3f472d9387b20a01b2
Binary files /dev/null and b/data/cache.dat differ
diff --git a/requirements.txt b/requirements.txt
index e5ec021d00fd1639e5e42286306bf772657433ca..04660f07555e3cd7644c4ba67b2ad04042d9650f 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -3,4 +3,3 @@ scipy==1.4.1
pandas==0.24.2
scikit-learn==0.22.1
joblib==0.14.1
-diskcache==4.1.0
diff --git a/tcmi.ipynb b/tcmi.ipynb
index ab6acac4e3ecd3e35c9b828f9ddb384f45979b67..cc16309926d6eca564523e784760c7117a6af83e 100644
--- a/tcmi.ipynb
+++ b/tcmi.ipynb
@@ -25,7 +25,7 @@
" Matthias Scheffler<sup>1</sup>,\n",
" and Luca Ghiringhelli<sup> 1</sup> <br><br>\n",
"<sup>1</sup> Fritz Haber Institute of the Max Planck Society, Faradayweg 4-6, D-14195 Berlin, Germany <br>\n",
- "<span class=\"nomad--last-updated\" data-version=\"v1.0.0\">[Last updated: January 14, 2020]</span>\n",
+ "<span class=\"nomad--last-updated\" data-version=\"v1.0.1\">[Last updated: January 17, 2020]</span>\n",
"</p>\n",
" \n",
"<div> \n",
@@ -75,8 +75,8 @@
"execution_count": 1,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T17:27:29.160459Z",
- "start_time": "2020-01-14T17:27:29.158391Z"
+ "end_time": "2020-01-17T09:14:38.542016Z",
+ "start_time": "2020-01-17T09:14:38.539754Z"
},
"init_cell": true
},
@@ -88,19 +88,22 @@
},
{
"cell_type": "markdown",
- "metadata": {},
+ "metadata": {
+ "heading_collapsed": true
+ },
"source": [
"## Import modules"
]
},
{
"cell_type": "code",
- "execution_count": 3,
+ "execution_count": 2,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T17:28:05.660112Z",
- "start_time": "2020-01-14T17:28:05.535093Z"
+ "end_time": "2020-01-17T09:14:39.810527Z",
+ "start_time": "2020-01-17T09:14:38.543402Z"
},
+ "hidden": true,
"init_cell": true
},
"outputs": [],
@@ -112,7 +115,6 @@
"import warnings\n",
"import functools\n",
"import itertools\n",
- "import diskcache\n",
"\n",
"import numpy as np\n",
"import scipy as sp\n",
@@ -124,6 +126,7 @@
"import tcmi\n",
"from tcmi import utils\n",
"from tcmi import entropy\n",
+ "from tcmi.cache import Cache\n",
"from tcmi.subspace_search import get_subspaces\n",
"from tcmi.estimators import DependenceEstimator\n",
"\n",
@@ -137,7 +140,7 @@
"# Main loop\n",
"if __name__ == '__main__': \n",
" # Provide cache\n",
- " storage = diskcache.Cache('.cache')\n",
+ " storage = Cache('data')\n",
" \n",
" # Configure plot environment\n",
" mpl.rc('font', family='sans', size=14)\n",
@@ -159,11 +162,7 @@
" \n",
" # General settings (please do not touch)\n",
" kwargs = dict(n_jobs=-1, return_scores=True)\n",
- " seed = 2019\n",
- "\n",
- "\n",
- " directory = 'thumbnails'\n",
- " os.makedirs(directory, exist_ok=True)"
+ " seed = 2019"
]
},
{
@@ -198,8 +197,8 @@
"execution_count": 5,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:49.116976Z",
- "start_time": "2020-01-14T16:57:49.114571Z"
+ "end_time": "2020-01-17T09:14:10.654185Z",
+ "start_time": "2020-01-17T09:14:10.651576Z"
},
"hidden": true
},
@@ -244,8 +243,8 @@
"execution_count": 6,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:49.154114Z",
- "start_time": "2020-01-14T16:57:49.118327Z"
+ "end_time": "2020-01-17T09:14:10.688047Z",
+ "start_time": "2020-01-17T09:14:10.655785Z"
},
"hidden": true
},
@@ -515,8 +514,8 @@
"execution_count": 7,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:49.581396Z",
- "start_time": "2020-01-14T16:57:49.155732Z"
+ "end_time": "2020-01-17T09:14:11.115344Z",
+ "start_time": "2020-01-17T09:14:10.689464Z"
},
"hidden": true,
"hide_input": false
@@ -702,8 +701,8 @@
"execution_count": 8,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:49.606945Z",
- "start_time": "2020-01-14T16:57:49.582702Z"
+ "end_time": "2020-01-17T09:14:11.136801Z",
+ "start_time": "2020-01-17T09:14:11.116550Z"
},
"hidden": true,
"scrolled": true
@@ -796,8 +795,8 @@
"execution_count": 9,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:50.149158Z",
- "start_time": "2020-01-14T16:57:49.608329Z"
+ "end_time": "2020-01-17T09:14:11.659328Z",
+ "start_time": "2020-01-17T09:14:11.138689Z"
},
"hidden": true
},
@@ -932,6 +931,7 @@
{
"cell_type": "markdown",
"metadata": {
+ "heading_collapsed": true,
"hidden": true
},
"source": [
@@ -953,8 +953,8 @@
"execution_count": 10,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:52.376153Z",
- "start_time": "2020-01-14T16:57:50.150470Z"
+ "end_time": "2020-01-17T09:14:14.090857Z",
+ "start_time": "2020-01-17T09:14:11.660643Z"
},
"hidden": true
},
@@ -964,11 +964,11 @@
"output_type": "stream",
"text": [
"Subspace: {} -> [-inf +- 0.00]\n",
- "x1 [Score: 0.29 +- 0.00]\n",
- "x2 [Score: 0.29 +- 0.00]\n",
- "x3 [Score: 0.29 +- 0.00]\n",
+ "x1 [Score: 0.22 +- 0.00]\n",
+ "x2 [Score: 0.22 +- 0.00]\n",
+ "x3 [Score: 0.22 +- 0.00]\n",
"Bound: -inf -> -inf\n",
- "[2020-01-14 17:57:52 - 0 subspaces remaining]\n",
+ "[2020-01-17 10:14:14 - 0 subspaces remaining]\n",
"\n",
"\n",
"Balance search tree\n"
@@ -1013,8 +1013,8 @@
"execution_count": 11,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:52.381592Z",
- "start_time": "2020-01-14T16:57:52.378442Z"
+ "end_time": "2020-01-17T09:14:14.095658Z",
+ "start_time": "2020-01-17T09:14:14.092601Z"
},
"hidden": true
},
@@ -1047,8 +1047,8 @@
"execution_count": 12,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:52.399855Z",
- "start_time": "2020-01-14T16:57:52.383463Z"
+ "end_time": "2020-01-17T09:14:14.117423Z",
+ "start_time": "2020-01-17T09:14:14.096814Z"
},
"hidden": true
},
@@ -1110,8 +1110,8 @@
"execution_count": 13,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:52.407133Z",
- "start_time": "2020-01-14T16:57:52.401094Z"
+ "end_time": "2020-01-17T09:14:14.125156Z",
+ "start_time": "2020-01-17T09:14:14.118966Z"
},
"hidden": true
},
@@ -1183,8 +1183,8 @@
"execution_count": 14,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:52.422195Z",
- "start_time": "2020-01-14T16:57:52.408387Z"
+ "end_time": "2020-01-17T09:14:14.139619Z",
+ "start_time": "2020-01-17T09:14:14.127330Z"
},
"hidden": true
},
@@ -1313,8 +1313,8 @@
"execution_count": 15,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:52.453354Z",
- "start_time": "2020-01-14T16:57:52.423327Z"
+ "end_time": "2020-01-17T09:14:14.165781Z",
+ "start_time": "2020-01-17T09:14:14.141425Z"
},
"hidden": true,
"scrolled": true
@@ -1541,8 +1541,8 @@
"execution_count": 16,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:52.521337Z",
- "start_time": "2020-01-14T16:57:52.454571Z"
+ "end_time": "2020-01-17T09:14:14.233510Z",
+ "start_time": "2020-01-17T09:14:14.167295Z"
},
"code_folding": [
0,
@@ -1764,8 +1764,8 @@
"execution_count": 17,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:53.012686Z",
- "start_time": "2020-01-14T16:57:52.522645Z"
+ "end_time": "2020-01-17T09:14:14.715484Z",
+ "start_time": "2020-01-17T09:14:14.234956Z"
},
"code_folding": [],
"hidden": true
@@ -1937,8 +1937,8 @@
"execution_count": 18,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:53.016279Z",
- "start_time": "2020-01-14T16:57:53.013963Z"
+ "end_time": "2020-01-17T09:14:14.718967Z",
+ "start_time": "2020-01-17T09:14:14.716748Z"
},
"hidden": true
},
@@ -1962,8 +1962,8 @@
"execution_count": 19,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:53.042121Z",
- "start_time": "2020-01-14T16:57:53.017442Z"
+ "end_time": "2020-01-17T09:14:14.745691Z",
+ "start_time": "2020-01-17T09:14:14.720322Z"
},
"hidden": true
},
@@ -2051,8 +2051,8 @@
"execution_count": 20,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:53.705286Z",
- "start_time": "2020-01-14T16:57:53.043392Z"
+ "end_time": "2020-01-17T09:14:14.834041Z",
+ "start_time": "2020-01-17T09:14:14.747010Z"
},
"hidden": true,
"scrolled": true
@@ -2465,8 +2465,8 @@
"execution_count": 21,
"metadata": {
"ExecuteTime": {
- "end_time": "2020-01-14T16:57:53.753938Z",
- "start_time": "2020-01-14T16:57:53.706466Z"
+ "end_time": "2020-01-17T09:14:14.881245Z",
+ "start_time": "2020-01-17T09:14:14.835250Z"
},
"scrolled": true
},
diff --git a/tcmi/cache.py b/tcmi/cache.py
new file mode 100644
index 0000000000000000000000000000000000000000..dd6d8664f4374ca67712802991483a0b23c02d79
--- /dev/null
+++ b/tcmi/cache.py
@@ -0,0 +1,100 @@
+# -*- coding: utf-8 -*-
+"""
+@package tcmi.cache
+
+@copyright Copyright (c) 2018+ Fritz Haber Institute of the Max Planck Society,
+ Benjamin Regler <regler@fhi-berlin.mpg.de>
+@license See LICENSE file for details.
+
+Licensed under the Apache License, Version 2.0 (the "License").
+You may not use this file except in compliance with the License.
+"""
+
+import os
+import gzip
+import pickle
+import tempfile
+
+
+class Constant(tuple):
+ "Pretty display of immutable constant."
+ def __new__(cls, name):
+ return tuple.__new__(cls, (name,))
+
+ def __repr__(self):
+ return '%s' % self[0]
+
+
+class Cache(object):
+ """Simple read-only disk-cache.
+ """
+
+ ENOVAL = Constant('ENOVAL')
+
+ def __init__(self, directory=None):
+ """Initialize cache instance.
+ """
+
+ if directory is None:
+ directory = tempfile.mkdtemp(prefix='diskcache-')
+ directory = os.path.expanduser(directory)
+ directory = os.path.expandvars(directory)
+
+ self._directory = directory
+ self._filename = os.path.join(directory, 'cache.dat')
+ if not os.path.isdir(directory):
+ try:
+ os.makedirs(directory, 0o755)
+ except OSError as error:
+ if error.errno != errno.EEXIST:
+ raise RuntimeError(
+ error.errno,
+ 'Cache directory "%s" does not exist'
+ ' and could not be created' % self._directory
+ )
+
+ data = {}
+ if os.path.exists(self._filename):
+ with gzip.open(self._filename, 'rb') as file:
+ data = pickle.load(file)
+ self._data = data
+
+ @property
+ def directory(self):
+ """Cache directory."""
+ return self._directory
+
+ @property
+ def filename(self):
+ """Cache filename."""
+ return self._filename
+
+ def __getitem__(self, key):
+ """Return corresponding value for `key` from cache.
+ """
+ value = self.get(key, default=self.ENOVAL)
+ if value is self.ENOVAL:
+ raise KeyError(key)
+ return value
+
+ def __setitem__(self, key, value):
+ """Set corresponding `value` for `key` in cache.
+ """
+ self.set(key, value)
+
+ def get(self, key, default=None):
+ """Retrieve value from cache. If `key` is missing, return `default`.
+ """
+ return self._data.get(key, default)
+
+ def set(self, key, value):
+ """Set `key` and `value` item in cache.
+ """
+ self._data[key] = value
+ return True
+
+ def save(self):
+ """Save cache to file.
+ """
+ with gzip.open(self._filename, 'wb') as file:
+ pickle.dump(self._data, file, pickle.HIGHEST_PROTOCOL)