CHANGELOG.rst 6.43 KB
Newer Older
Jason Zhang's avatar
Jason Zhang committed
1 2 3 4 5 6
v0.3.3.1 (2018-02-10)
---------------------

* Fixed bug in which MarkovChain.steady_state might use the wrong eigenvector.
* Added warnings if transition probabilities are negative or don't sum to 1.

Jason Zhang's avatar
Jason Zhang committed
7 8 9 10 11 12 13
v0.3.3.0 (2018-02-08)
---------------------

* plot_path now only plots the path if the probability is nonzero
* plot_path takes in a starting condition to be consistent with prob_of_path
* simulate_path takes an optional parameter to plot the path

14 15 16 17 18
v0.3.2.1 (2018-02-08)
---------------------

* Only show real component of steady state distribution when computing left eigenvector.

19 20 21
v0.3.2.0 (2018-01-18)
---------------------

22
* Restore functionality for calling `probability_function` for joint distributions.
23 24


Jason Zhang's avatar
Jason Zhang committed
25 26 27 28 29 30 31
v0.3.1.1 (2018-01-08)
---------------------

* MarkovChain.distribution() supports a state as a starting distribution.

v0.3.1.0 (2018-01-08)
---------------------
32 33

* Code refactor to follow PEP8
Jason Zhang's avatar
Jason Zhang committed
34
* All new Markov Chain module using numpy backend
35 36

  * Function definitions largely the same for common functions
37
  * MarkovChains can now be constructed using additional class functions
38

Jason Zhang's avatar
Jason Zhang committed
39
* Functions renamed:
40 41 42 43

  * Table.expected_value -> Table.ev
  * Table.variance -> Table.var

44
* Unit Tests! Bumped to around 66% code coverage.
45

Jason Zhang's avatar
Jason Zhang committed
46 47 48 49 50
v0.2.9.0 (2017-03-19)
---------------------

* Plot_3d

51 52 53 54 55 56
v0.2.8.1 (2017-03-18)
---------------------

* Plot_continuous now accepts python functions too


57 58 59 60 61
v0.2.8.0 (2017-03-13)
---------------------

* Updated unconstrain to rearrange_2 and nicefy to rearrange_1

Dibya Ghosh's avatar
Dibya Ghosh committed
62
v0.2.7.1 (2017-03-11)
63 64 65 66
---------------------

* SymPy integration being finalized - added `unconstrain` and updated `declare`

Jason Zhang's avatar
Jason Zhang committed
67 68 69 70 71
v0.2.7.0 (2017-03-10)
---------------------

* sample renamed to sample_from_dist to avoid conflicts with datascience

Jason Zhang's avatar
Jason Zhang committed
72 73 74 75 76 77
v0.2.6.3 (2017-03-09)
---------------------

* Fixed documentation for plots
* plots removed from global

Jason Zhang's avatar
Jason Zhang committed
78 79 80 81 82
v0.2.6.2 (2017-03-09)
---------------------

* Plot_continuous works with sympy

83 84 85 86 87
v0.2.6.1 (2017-03-09)
---------------------

* Plot_continuous now works with any function passed in as func

Jason Zhang's avatar
Jason Zhang committed
88 89 90 91 92
v0.2.6.0 (2017-03-06)
---------------------

* Wrapper for plotting continuous functions

Dibya Ghosh's avatar
Dibya Ghosh committed
93 94 95 96 97
v0.2.5.1 (2017-03-06)
---------------------

* Beginning to add SymPy integration in *symbolic_math.py*

Jason Zhang's avatar
Jason Zhang committed
98 99 100 101 102
v0.2.5.0 (2017-02-22)
---------------------

* Added log_probability_of_path

Jason Zhang's avatar
Jason Zhang committed
103
v0.2.4.4 (2017-02-20)
Jason Zhang's avatar
Jason Zhang committed
104 105
---------------------

Jason Zhang's avatar
Jason Zhang committed
106 107 108 109 110
* Fixing installation issues

v0.2.4.3a (2017-02-20)
----------------------

Jason Zhang's avatar
Jason Zhang committed
111
* fixed mfpt
Jason Zhang's avatar
Jason Zhang committed
112

Jason Zhang's avatar
Jason Zhang committed
113 114 115 116 117 118 119 120 121 122
v0.2.4.2 (2017-02-16)
---------------------

* Fixed typo in steady_state, not sure how it happened

v0.2.4.1 (2017-02-16)
---------------------

* Documentation fix

123 124 125 126 127 128 129
v0.2.4.0 (2017-02-13)
---------------------

* Removed T and S from markov chains
* added .column
* states now sorted

Jason Zhang's avatar
Jason Zhang committed
130 131 132 133 134
v0.2.3.8 (2017-02-13)
---------------------

* Added get target

135 136 137 138 139 140 141 142 143 144
v0.2.3.7 (2017-02-12)
---------------------

* Deprecation error fix

v0.2.3.6 (2017-02-12)
---------------------

* Distribution now shows states with probability 0

145 146 147 148 149
v0.2.3.5 (2017-02-11)
---------------------

* Added show_ev for conditional distributions

Jason Zhang's avatar
Jason Zhang committed
150 151 152 153 154
v0.2.3.4 (2017-02-11)
---------------------

* state --> states

Jason Zhang's avatar
Jason Zhang committed
155 156
v0.2.3.3
--------
Jason Zhang's avatar
Jason Zhang committed
157
* Documentation
Jason Zhang's avatar
Jason Zhang committed
158 159 160 161 162

v0.2.3.2 (2017-02-11)
---------------------
* Changed label for empirical distribution to state
* mc.distribution accepts states
Jason Zhang's avatar
Jason Zhang committed
163

Jason Zhang's avatar
Jason Zhang committed
164 165 166
v0.2.3.1 (2017-02-11)
---------------------

167 168 169 170 171
* Fixed mean_first_passage_times

v0.2.3.0 (2017-02-11)
---------------------

Jason Zhang's avatar
Jason Zhang committed
172 173 174 175 176 177 178 179
* Renamed a ton of functions
* Implemented starting conditions

v0.2.2.0 (2017-02-11)
---------------------

* Begin wrapping of pykov

Jason Zhang's avatar
Jason Zhang committed
180 181 182 183 184 185 186
v0.2.1.3 (2017-02-08)
---------------------

* Plots uses plt.bar instead of Table.hist
* Added optional parameter edges=


Jason Zhang's avatar
Jason Zhang committed
187 188 189 190 191
v0.2.1.2 (2017-02-04)
---------------------

* Added show_ave as optional parameter

Jason Zhang's avatar
Jason Zhang committed
192 193 194
v0.2.1.1 (2017-02-04)
---------------------

Jason Zhang's avatar
Jason Zhang committed
195
* Added show_ev and show_sd as optional parameters for plot
Jason Zhang's avatar
Jason Zhang committed
196 197 198 199 200 201 202

v0.2.1.0 (2017-02-04)
---------------------

* Added sample for single variable distributions
* Added CDF for single variable distributions

Jason Zhang's avatar
Jason Zhang committed
203 204 205 206 207
v0.2.0.0 (2017-02-03)
---------------------

* Pykov

Jason Zhang's avatar
Jason Zhang committed
208 209 210 211 212
v0.1.8.1 (2017-02-01)
---------------------

* Renamed emp_dist values to proportions rather than probabilities

Jason Zhang's avatar
Jason Zhang committed
213 214 215 216 217 218 219 220 221 222 223
v0.1.8.0 (2017-01-30)
---------------------

* Added emp_dist to allow for empirical distributions


v0.1.7.6 (2017-01-19)
---------------------

* __version__ instead of version

Jason Zhang's avatar
Jason Zhang committed
224 225 226 227 228
v0.1.7.5 (2017-01-18)
---------------------

* Joint Distributions no longer give a warning if probabilities rounded to 6 decimal places = 1

229 230 231 232 233
v0.1.7.4 (2017-01-17)
---------------------

* Single variable distributions now check that probabilities sum to 1

234 235 236 237 238 239 240 241 242 243 244 245
v0.1.7.3 (2017-01-17)
---------------------

* Plot now adds edge border if there are fewer than 75 bins
* Plot now has an optional parameter edge that accepts a boolean
* Added marginal_dist which returns a single variable distribution

v0.1.7.2 (2017-01-17)
---------------------

* .values is now an alias for .domain

Jason Zhang's avatar
Jason Zhang committed
246 247 248 249 250
v0.1.7.1 (2017-01-17)
---------------------

* Fixed vertical axis for Plot

251 252 253 254 255 256
v0.1.7.0 (2017-01-16)
---------------------

* Removed marginal_of_X, marginal_of_Y, etc
* conditional_dist_given(given) is now conditional_dist(label, given)

Jason Zhang's avatar
0.1.6.4  
Jason Zhang committed
257 258 259 260 261
v0.1.6.4 (2017-01-15)
---------------------

* Joint Distribution functions can have arbitrary number of arguments again

Jason Zhang's avatar
Jason Zhang committed
262 263 264 265 266 267 268 269 270 271 272 273 274 275 276
v0.1.6.3 (2017-01-15)
---------------------

* fixed a bug in which toJoint just renamed the x-columns rather than changing the order

v0.1.6.2 (2017-01-14)
---------------------

* toJoint now preserve original order

v0.1.6.1 (2017-01-14)
---------------------

* JointDistribution probabilities don't have to sum to 1,

Jason Zhang's avatar
v0.1.6  
Jason Zhang committed
277 278 279 280 281 282
v0.1.6 (2017-01-14)
-------------------

* Added probability_function for JointDistribution
* probability_function now checks number of arguments in pfunc

Jason Zhang's avatar
Jason Zhang committed
283 284 285 286 287 288 289 290 291 292 293
v0.1.5.1 (2017-01-12)
---------------------

* Added JointDistribution to the init

v0.1.5 (2017-01-12)
-------------------

* Plotting width now works with events and masks
* JointDistribution can now be used with any variable

Jason Zhang's avatar
Jason Zhang committed
294 295 296 297 298 299 300 301 302 303
v0.1.4.3 (2016-12-20)
---------------------

* Changed the colors for plots

v0.1.4.2
--------

* Slight modifications to plot labels

Jason Zhang's avatar
Jason Zhang committed
304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320
v0.1.4a
-------

* Single distribution plotting moved from the ``plot_dist`` method to the ``Plot`` function
* Multiple distribution plotting moved from the ``Plot`` function to the ``Plots`` function
* Events are now plotted by passing an argument to ``Plot``

v0.1.3
------

* Added joint distributions
* All ``FiniteDistribution`` objects changed to become ``datascience.tables.Table`` objects
* Began renaming

v0.1.2
------
Initial Release