Commit Graph

  • 9ffe9303b3 Merge pull request #50 from BradyBromley/master Aurélien Geron 2019-10-27 06:56:15 +0800
  • 63978befe8 Changed wording for clarity in README.md BradyBromley 2019-10-25 10:03:33 -0700
  • bbadc3276a Merge branch 'master' of github.com:ageron/handson-ml2 Aurélien Geron 2019-10-21 09:02:13 +0800
  • 5edb3f3e51 Add protoc example and tf.io.decode_proto() example Aurélien Geron 2019-10-21 09:01:15 +0800
  • 143ef4b1db Remove y=None argument in the transform() method Aurélien Geron 2019-10-17 18:05:01 +0800
  • 53dc780956 Merge pull request #41 from jmribeiro/master Aurélien Geron 2019-10-14 10:11:13 +0800
  • aed738c7ad Removed duplicated np.random.seed(42) entry from the second notebook jmribeiro 2019-10-13 19:20:42 +0100
  • b5cbdc9276 Add environment.yml for conda, and update Dockerfile to use it Aurélien Geron 2019-10-13 21:56:08 +0800
  • 89653bfade Improve a few figures (e.g., add missing labels, share axes, etc.) Aurélien Geron 2019-10-13 17:19:39 +0800
  • 7e6489f8a4 Improve a few figures (e.g., add missing labels, share axes, etc.) Aurélien Geron 2019-10-13 16:58:36 +0800
  • 23b6366c39 Update to TF 2.0.0 (from nightly) and update other libraries Aurélien Geron 2019-10-13 00:19:05 +0930
  • d8971f1767 Fix os.join() => os.path.join() Aurélien Geron 2019-10-12 18:05:41 +0930
  • 2913a17508 Work around "import tensorflow.train" bug. Aurélien Geron 2019-10-12 17:42:41 +0930
  • 5c9957065b Work around "import tensorflow.train" bug. Aurélien Geron 2019-10-12 17:38:40 +0930
  • 02205e034f Remove from __future__ imports as we move away from Python 2 Aurélien Geron 2019-10-12 17:07:53 +0930
  • 7d0052f146 Import urllib directly instead of from six.moves, as we move away from Python 2 Aurélien Geron 2019-10-12 16:42:23 +0930
  • f6dfa0ff76 Remove from __future__ imports as we move away from Python 2 Aurélien Geron 2019-10-12 16:40:05 +0930
  • 936e2cf50f Import urllib directly instead of from six.moves, as we move away from Python 2 Aurélien Geron 2019-10-12 16:30:13 +0930
  • 07b83a5e09 Import urllib directly instead of from six.moves, as we move away from Python 2 Aurélien Geron 2019-10-12 16:29:54 +0930
  • f5f1ce1ca0 Remove deepnote note about startup speed Aurélien Geron 2019-10-01 09:51:50 +0930
  • 3f634d9eb9 Merge pull request #35 from Jirka-Lhotka/patch-1 Aurélien Geron 2019-10-01 09:48:49 +0930
  • 41686e16d2 Update Deepnote's note Jirka-Lhotka 2019-09-30 13:58:49 +0200
  • f798a5c932 Merge pull request #34 from thierryherrmann/no_dropout_in_stateful_rnn Aurélien Geron 2019-09-30 15:30:10 +0930
  • 22aa520ccc issue 32: remove dropout in stateful RNN Thierry Herrmann 2019-09-28 12:06:05 -0400
  • b9509ced94 First to do => First thing to do Aurélien Geron 2019-09-28 22:55:02 +0800
  • 568aea61e0 Remove {nbsp} twice Aurélien Geron 2019-09-02 21:35:33 +0800
  • feb1dcf4a6 Merge pull request #26 from thierryherrmann/chap09_likelihood_fix Aurélien Geron 2019-09-02 15:46:46 +0800
  • 2c1c67e89a Replace 1^st^ with 1st. Aurélien Geron 2019-09-02 15:22:35 +0800
  • 497ec6d2aa chap09 likelihood fix Thierry Herrmann 2019-09-01 12:55:16 -0400
  • 15b9e4507e Add a doc describing the changes in the 2nd edition Aurélien Geron 2019-08-27 16:18:40 +0800
  • 0c4a9820fd Update to nltk 3.4.5 (security patch) Aurélien Geron 2019-08-27 13:26:02 +0800
  • ea3d17c063 Fix execution_counts Aurélien Geron 2019-08-12 14:51:06 +0800
  • 87ea670bef Iris-Virginica => Iris virginica (& same for versicolor and setosa) Aurélien Geron 2019-08-12 14:45:33 +0800
  • aad780b74b Fix Actual vs Forecast in diagrams, fixes #21 Aurélien Geron 2019-08-03 12:11:40 +0800
  • 281366776f Add 19_training_and_deploying_at_scale.ipynb, update index.ipynb Aurélien Geron 2019-07-13 11:41:30 +0200
  • 8f7a2c355b Display Fashion MNIST Aurélien Geron 2019-07-13 11:31:17 +0200
  • 6c0cd5d2df Improve the code to forecast the next 10 time steps Aurélien Geron 2019-07-13 11:28:50 +0200
  • f8f2b9e4bb Add (USD) after GDP per capita Aurélien Geron 2019-07-10 17:08:12 +0200
  • 48dfa2ce67 tf.image.resize_image_with_crop_or_pad was renamed to tf.image.resize_with_crop_or_pad Aurélien Geron 2019-06-27 10:10:22 +0200
  • abf3bba2d5 Clarify DenseTranspose Aurélien Geron 2019-06-10 17:42:31 +0800
  • 400920f0aa SGD now defaults to lr=0.01 so use 1e-3 explicitly Aurélien Geron 2019-06-10 10:53:32 +0800
  • 3db31444cd SGD now defaults to lr=0.01 so use 1e-3 explicitely Aurélien Geron 2019-06-10 10:48:00 +0800
  • 1973371b19 Use causal padding in WaveNet and fix forecast Aurélien Geron 2019-06-09 23:56:19 +0800
  • 01784d2f98 Do not use a layer as an activation function, especially with weights Aurélien Geron 2019-06-09 20:08:53 +0800
  • 0023162fc5 bst_n_estimators should be argmin + 1 Aurélien Geron 2019-06-08 21:59:55 +0800
  • e9ae4997a5 Update requirements, e.g. add tf-agents, tf serving api, etc. Aurélien Geron 2019-06-08 12:01:58 +0800
  • b6306b7d48 Update ch. 10 code to latest TF nightly version, eg. default SGD learning rate is now 0.001 Aurélien Geron 2019-06-08 11:43:48 +0800
  • 0615a6f0a1 LayerNormalization is not experimental anymore Aurélien Geron 2019-06-06 11:06:13 +0800
  • fb47fab0c8 Update docker image for 2nd edition Aurélien Geron 2019-06-04 16:43:14 +0800
  • f680e49ea2 Replace SGD with SVC in OvA vs OvO section Aurélien Geron 2019-05-28 15:21:49 +0800
  • efa16f02ec Add breakout.gif Aurélien Geron 2019-05-28 09:31:22 +0800
  • 4c3b7b9b06 Save agent's breakout performance to an animated gif Aurélien Geron 2019-05-28 09:30:16 +0800
  • 3ef350ab4c Fix figure name and clarify a couple code examples Aurélien Geron 2019-05-27 20:35:00 +0800
  • c5f4b41cf5 Fix breakout plot Aurélien Geron 2019-05-26 23:56:49 +0800
  • 2edbb6e9d4 Add Reinforcement Learning notebook Aurélien Geron 2019-05-26 23:30:39 +0800
  • 80279c4477 Make notebook 09 deterministic, and avoid non-convergence warnings Aurélien Geron 2019-05-19 16:19:29 +0700
  • c702bfbcab Fix ladybug.png so imageio can load it without chunk_eXIf issues Aurélien Geron 2019-05-19 15:03:42 +0700
  • 73a36f335f Add warning about TF issue regarding DenseFeatures and the Functional API, fixes #6 Aurélien Geron 2019-05-15 20:23:24 +0800
  • ec3136f59f Add index=housing.index when wrapping array in a DataFrame, fixes #426 Aurélien Geron 2019-05-12 21:28:56 +0800
  • f0f83903bd Fix the transformer (use final encoder outputs) Aurélien Geron 2019-05-10 21:30:18 +0800
  • 7b1c890195 Stop using learning_phase: it's a global variable, and currently broken anyway Aurélien Geron 2019-05-09 16:26:38 +0800
  • a758ea50b0 Rename the figure paths from *_diagram.png to *_plot.png Aurélien Geron 2019-05-09 16:25:40 +0800
  • 8e08319781 Rename the figure paths from *_diagram.png to *_plot.png Aurélien Geron 2019-05-09 16:25:26 +0800
  • d4d15a9c59 Clarify figure 4-19 Aurélien Geron 2019-05-09 16:24:17 +0800
  • 107b90d9b3 Do not use learning_phase anymore, just set training=True/False Aurélien Geron 2019-05-09 10:39:02 +0800
  • df100ffa9a Rename figure Aurélien Geron 2019-05-06 13:15:15 +0800
  • dd5e648c62 Rename figure Aurélien Geron 2019-05-06 13:15:01 +0800
  • 22eb374304 Rename figure Aurélien Geron 2019-05-06 13:14:50 +0800
  • 4941749b60 Fix typo in figure name Aurélien Geron 2019-05-06 13:14:23 +0800
  • aa648067a9 Change notebook title to Autoencoders and GANs Aurélien Geron 2019-05-06 09:38:47 +0800
  • 4e155bc23b Rename 17 to Autoencoders and GANs Aurélien Geron 2019-05-06 09:37:56 +0800
  • 42f24fa44f Add GANs Aurélien Geron 2019-05-06 09:34:45 +0800
  • fd1e088dab Add 1cycle scheduling Aurélien Geron 2019-05-05 12:42:08 +0800
  • d2a518cdb1 Add more chapters to index.ipynb Aurélien Geron 2019-04-30 15:01:15 +0800
  • 7335c9d5c0 Use separate joblib package rather than the one in sklearn (which is deprecated) Aurélien Geron 2019-04-26 21:22:15 +0800
  • 327407501e Add notebook 17 on autoencoders Aurélien Geron 2019-04-26 21:19:32 +0800
  • 9925735995 Add the images/nlp directory Aurélien Geron 2019-04-23 20:06:24 +0800
  • 6f98bc5a04 Stateful RNNs now support recurrent_dropout Aurélien Geron 2019-04-21 15:55:39 +0800
  • 33d1894dc1 Update requirements to use TensorFlow 2.0 alpha0 Aurélien Geron 2019-04-20 13:05:52 +0800
  • df9b91e2e8 Rename chapters 11 to 15 and split chapter 15 into 15 and 16 Aurélien Geron 2019-04-16 20:39:14 +0800
  • 61bb73074b Add a basic implementation of MultiHeadAttention Aurélien Geron 2019-04-16 19:52:49 +0800
  • d24fee1576 Add .bak.* and datasets/titanic to .gitignore Aurélien Geron 2019-04-16 08:44:30 +0800
  • 03810ea2f7 Add chapter 15, time series and NLP using RNNs, CNNs and Attention Aurélien Geron 2019-04-16 08:27:36 +0800
  • 03c5ef896b Add chapter 15 - Handling sequential data (time series and NLP with RNNs and Attention) Aurélien Geron 2019-04-16 00:09:10 +0800
  • 52afcf4290 Crop long outputs so they show up nicer on github.com Aurélien Geron 2019-04-16 00:06:57 +0800
  • 687fd71b5f Add outputs for threshold_90_precision and fix comment, fixes #2 Aurélien Geron 2019-04-05 17:06:37 +0800
  • 6d7dcd7c80 Work in progress updating ch15 Aurélien Geron 2019-04-05 17:04:38 +0800
  • 6974508cf7 Remove old TensorFlow chapter Aurélien Geron 2019-03-25 18:14:29 +0800
  • f0b432eb59 Add missing BN layer in ResNet-34 and remove bias in Conv2D Aurélien Geron 2019-03-25 12:19:27 +0800
  • 3313c21723 No bias before BN layers Aurélien Geron 2019-03-25 12:03:44 +0800
  • 1774ede668 Update CNN chapter Aurélien Geron 2019-03-24 09:06:29 +0800
  • 7d8dfb45cf Move CNN chapter out of work_in_progress Aurélien Geron 2019-03-16 13:59:41 +0800
  • 5f6be6fa6f Use pd.cut() for income_cat Aurélien Geron 2019-03-15 23:49:03 +0800
  • e8ed65eac6 Update index.ipynb with chapters 8 to 13 Aurélien Geron 2019-03-15 23:41:04 +0800
  • 2f78d1f1e5 Add tf datasets, tf hub, tf probability and tf transform to requirements Aurélien Geron 2019-03-15 12:07:50 +0800
  • 990fb1a453 Add notebook for chapter 13, loading and preprocessing data Aurélien Geron 2019-03-14 09:15:09 +0800
  • af3f11dbc9 Add chapter 12 notebook Aurélien Geron 2019-03-05 19:46:03 +0800
  • e938df8fb9 Add kwargs to constructor of custom model Aurélien Geron 2019-02-28 19:48:50 +0800
  • ed5c7d6db8 Add MC Dropout, finish chapter Aurélien Geron 2019-02-28 19:48:06 +0800
  • 6bc603b508 Move 13_deep_learning.ipynb to 11_deep_learning.ipynb Aurélien Geron 2019-02-17 20:31:59 +0800