schuetoliver
/
handson-ml
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
1,114 Commits
1 Branch
0 Tags
32 MiB
Commit Graph

1114 Commits (e0f7342083c81d9140922a2fa572b0a5157dfb65)

Author SHA1 Message Date
Aurélien Geron c224d8e74c Fix expected result for first estimator training 2018-09-03 23:22:17 +02:00
Aurélien Geron d956648eb4 Fix the virtualenv activation command for Windows (fixes #287) 2018-09-03 22:26:31 +02:00
Aurélien Geron 1c09995f1b Remove unnecessary variable initialization, fixes #283 2018-08-20 18:55:24 +02:00
Aurélien Geron bb1cc02950 Merge branch 'master' of github.com:ageron/handson-ml 2018-08-11 12:21:11 +01:00
Aurélien Geron bbeef54739 Add a notebook about achieving reproducibility with TensorFlow 2018-08-11 12:20:57 +01:00
Aurélien Geron c2952be7f2 Merge pull request #276 from dgwozdz/master 
Correction of a typo
 2018-08-07 16:08:50 +01:00
kaksat b9269e7207 Correction of a typo 
Current version produces the following error:

AttributeError: module 'matplotlib.cm' has no attribute 'spectral'
 2018-08-06 21:57:40 +02:00
Aurélien Geron d0cf5681c3 Merge pull request #275 from dgwozdz/master 
Correction of a formula for silhouette coefficient
 2018-08-06 19:53:57 +01:00
kaksat 50aeba2e67 Correction of a formula for silhouette coefficient 
Source:
http://scikit-learn.org/stable/modules/generated/sklearn.metrics.silhouette_score.html
 2018-08-06 20:21:35 +02:00
Aurélien Geron 2688da0477 Make notebook 14 reproducible (except for the dropout section, that one is tough), fixes #274 2018-08-06 16:28:10 +01:00
Aurélien Geron 56be8647f9 Get output of hidden3, not hidden4 (fixes #273). 2018-08-03 16:15:18 +01:00
Aurélien Geron a82e624171 Fix comment in DataFrameSelector 2018-07-31 23:10:23 +01:00
Aurélien Geron 9367c6ef42 Silence gelsd warning 2018-07-31 20:22:05 +01:00
Aurélien Geron de9f490bc3 Add ColumnTransformer to future_encoders.py 2018-07-31 20:09:12 +01:00
Aurélien Geron e2d450708a Replace FeatureUnion + DataFrameSelector with new ColumnTransformer 2018-07-31 20:08:33 +01:00
Aurélien Geron 060751a976 Fix my_reber_classifier path (fixes #267) 2018-07-30 18:58:02 +01:00
Aurélien Geron a28604d023 Fix probability at the end of the Reber exercise (fixes #266) 2018-07-30 18:52:50 +01:00
Aurélien Geron 0e0483d5ee Add missing training: True for dropout, and point to tf.nn.selu, fixes #228 2018-05-27 21:24:24 +02:00
Aurélien Geron a7c50e12c4 Add boxplot in notebook 03, exercise 3 2018-05-26 15:01:33 +02:00
Aurélien Geron 703815bca9 Use tensorflow_graph_in_jupyter.py in notebooks, fixes #223 2018-05-18 21:22:56 +02:00
Aurélien Geron ef2a7b5bd0 Add jupyter-tensorboard to requirements.txt, it's pretty handy 2018-05-18 20:53:38 +02:00
Aurélien Geron 21794905c6 Improve formatting 2018-05-18 17:17:51 +02:00
Aurélien Geron dc2b38fc83 Add a comment about why Deep Learning prefers reverse mode autodiff rather than forward mode 2018-05-18 17:09:59 +02:00
Aurélien Geron a64e492c6d Fix a typo in extra_autodiff 2018-05-18 17:03:46 +02:00
Aurélien Geron 60653bba3e Add tensorflow_graph_in_jupyter.py 2018-05-18 16:52:10 +02:00
Aurélien Geron 39c9821249 Move autodiff content to its own notebook extra_autodiff.ipynb 2018-05-18 16:47:20 +02:00
Aurélien Geron 3af8bcd1bb Update ReplayMemory class and add a short description 2018-05-09 16:54:17 +02:00
Aurélien Geron c6b424931a Merge pull request #204 from NileshPS/master 
Ch 16. Using native python list instead of deque for storing experience, increases sampling performance considerably.
 2018-05-09 15:34:12 +02:00
Aurélien Geron 10bc49aa91 Add solution to exercise 16.8 (BipedalWalker-v2 using Policy Gradients) 2018-05-09 15:25:49 +02:00
Aurélien Geron 4d25a60d6a Add solution of exercises 7.8 (voting classifier) and 7.9 (stacking ensemble) 2018-05-09 10:09:04 +02:00
Aurélien Geron cac02c5964 Updated to latest version of TensorFlow 2018-05-08 20:21:23 +02:00
Aurélien Geron 1886d796e1 Updated to latest version of TensorFlow 2018-05-08 20:16:00 +02:00
Aurélien Geron 55adea1ff4 Add code to compute a confidence interval 2018-05-08 19:41:47 +02:00
Aurélien Geron e05d4b36ac tf.contrib.layers.variance_scaling_initializer moved to tf.variance_scaling_initializer 2018-05-08 19:40:05 +02:00
Aurélien Geron 38c2ea79e4 Notebook 08 now has more unsupervised learning content (clustering, anomaly detection, density estimation, gaussian mixtures...) 2018-05-08 12:53:40 +02:00
Aurélien Geron f45a88146f Add instructions to install using the zip file rather than git, if necessary 2018-05-08 12:52:47 +02:00
Aurélien Geron 581253b47a Add xgboost example and upgrade to latest matplotlib version 2018-05-08 12:43:49 +02:00
Aurélien Geron d9fdb0542b Remove linewidth argument for plt.contourf() 2018-05-08 12:38:05 +02:00
Aurélien Geron fb3e68709e Replace n_iter with max_iter in SGDClassifier 2018-05-08 12:36:43 +02:00
Aurélien Geron c4e72ddc49 Replace pip with pip3, and add random_state to LogisticRegression 2018-05-08 12:33:46 +02:00
Aurélien Geron 4dbb575933 Merge pull request #161 from nbgraham/patch-2 
Ch. 3: Updates explanation to match data
 2018-05-08 11:25:19 +02:00
Aurélien Geron fb4e5aa4ea Fix equation 16-6 (max_alpha'=>max_a') 2018-05-07 22:47:28 +02:00
Aurélien Geron 2af3c59dd6 Add *.old, *.dot and lifesat.csv (generated) to .gitignore 2018-05-07 22:46:44 +02:00
Aurélien Geron 373a52c39e Add a quick intro to the Data API in notebook 12 2018-05-07 22:45:48 +02:00
Aurélien Geron 0a1b040d9d Add extra material in notebook 08: clustering, anomaly detection, gaussian mixtures 2018-05-07 21:20:49 +02:00
Aurélien Geron 771dccaca4 Clarify future encoders in Scikit-Learn 0.20 2018-05-07 21:09:08 +02:00
Aurélien Geron 77d3d4838d Clarify why we are using OrdinalEncoder and OneHotEncoder 2018-05-07 20:17:30 +02:00
Aurélien Geron 46f547daeb Replace the CategoricalEncoder with the OneHotEncoder class 2018-05-07 19:52:01 +02:00
Aurélien Geron bdd58b7c52 Improve the formatting of a few equations, in particular theta vector should have a bold font 2018-05-07 16:13:51 +02:00
Aurélien Geron 8fa23cceaf Clarify the mini-batch K-Means performance comparison diagram 2018-05-07 16:07:15 +02:00