Tensorflow

From Robin

(Difference between revisions)
Jump to: navigation, search
(Ny side: Tensorflow https://www.tensorflow.org is an open source library for defining and evaluating computational graphs. It’s great for developing Artificial Neural Networks and includes many u…)
Line 1: Line 1:
-
Tensorflow https://www.tensorflow.org  is an open source library for defining and evaluating computational graphs. It’s great for developing Artificial Neural Networks and includes many useful components (e.g., LSTM and various RNN cell designs) and optimisation tools (e.g., Adam).  
+
[Tensorflow https://www.tensorflow.org] is an open source library for defining and evaluating computational graphs. It’s great for developing Artificial Neural Networks and includes many useful components (e.g., LSTM and various RNN cell designs) and optimisation tools (e.g., Adam).  
-
Tensorflow is most useful if you want to work at the “operator” level of machine learning development, e.g., defining special kinds of output layers. If you want to work with existing layers, it may be better to use a higher-level library such as Keras: https://keras.io  
+
Tensorflow is most useful if you want to work at the ''operation'' level of machine learning development, e.g., defining special kinds of output layers. If you want to work with existing layers, it may be better to use a higher-level library such as [Keras: https://keras.io].
Tensorflow is still new (1.0 came out in February 2017) and can be a bit of a moving target (particularly the contrib parts), but it includes some great tools such as Tensorboard for visualisation and is good at computing things on GPUs.
Tensorflow is still new (1.0 came out in February 2017) and can be a bit of a moving target (particularly the contrib parts), but it includes some great tools such as Tensorboard for visualisation and is good at computing things on GPUs.
-
## Installing:
+
## Installing: ##
1. Install python of choice
1. Install python of choice
Line 12: Line 12:
4. $$!
4. $$!
-
## Resources:
+
## Resources: ##
Here’s some useful resources to get started:  
Here’s some useful resources to get started:  
 +
### Courses ###
-
### Courses
+
* [Udacity: Deep Learning with Google: https://www.udacity.com/course/deep-learning--ud730]
-
- Udacity: Deep Learning with Google: https://www.udacity.com/course/deep-learning--ud730
+
### Books ###
-
### Books
+
* [Hands-on Machine Learning with Scikit-Learn and Tensorflow http://shop.oreilly.com/product/0636920052289.do]. Probably the best current book on TF. Second half of the book is a great primer for DL with practical examples about implementing ideas in TensorFlow. Code examples are here: https://github.com/ageron/handson-ml
-
- Hands-on Machine Learning with Scikit-Learn and Tensorflow: http://shop.oreilly.com/product/0636920052289.do Probably the best current book on TF. Second half of the book is a great primer for DL with practical examples about implementing ideas in TensorFlow. Code examples are here: https://github.com/ageron/handson-ml
 
 +
### Blogs ###
-
### Blogs
+
* Practical Guide for Debugging TensorFlow Code: https://wookayin.github.io/tensorflow-talk-debugging/#1
 +
* Tensorflow Quick Tips: http://www.deeplearningweekly.com/blog/tensorflow-quick-tips
 +
* TF Best Practices: https://github.com/aicodes/tf-bestpractice
 +
* Structuring your TensorFlow models: http://danijar.com/structuring-your-tensorflow-models/
 +
* TensorBoard: Visualizing Learning: https://www.tensorflow.org/get_started/summaries_and_tensorboard
 +
* RNNs in TensorFlow: http://www.wildml.com/2016/08/rnns-in-tensorflow-a-practical-guide-and-undocumented-features/
 +
* RNNs in TensorFlow I: https://r2rt.com/recurrent-neural-networks-in-tensorflow-i.html and II: https://r2rt.com/recurrent-neural-networks-in-tensorflow-ii.html
-
- Practical Guide for Debugging TensorFlow Code: https://wookayin.github.io/tensorflow-talk-debugging/#1
+
### Examples: ###
-
- Tensorflow Quick Tips: http://www.deeplearningweekly.com/blog/tensorflow-quick-tips
+
-
- TF Best Practices: https://github.com/aicodes/tf-bestpractice
+
-
- Structuring your TensorFlow models: http://danijar.com/structuring-your-tensorflow-models/
+
-
- TensorBoard: Visualizing Learning: https://www.tensorflow.org/get_started/summaries_and_tensorboard
+
-
- RNNs in TensorFlow: http://www.wildml.com/2016/08/rnns-in-tensorflow-a-practical-guide-and-undocumented-features/
+
-
- RNNs in TensorFlow I: https://r2rt.com/recurrent-neural-networks-in-tensorflow-i.html and II: https://r2rt.com/recurrent-neural-networks-in-tensorflow-ii.html
+
-
### Examples:
+
* Char-rnn-tensorflow: https://github.com/sherjilozair/char-rnn-tensorflow  
-
 
+
* RNN TF Tutorial: https://github.com/silicon-valley-data-science/RNN-Tutorial
-
- Char-rnn-tensorflow: https://github.com/sherjilozair/char-rnn-tensorflow  
+
-
- RNN TF Tutorial: https://github.com/silicon-valley-data-science/RNN-Tutorial
+

Revision as of 09:19, 12 June 2017

[Tensorflow https://www.tensorflow.org] is an open source library for defining and evaluating computational graphs. It’s great for developing Artificial Neural Networks and includes many useful components (e.g., LSTM and various RNN cell designs) and optimisation tools (e.g., Adam).

Tensorflow is most useful if you want to work at the operation level of machine learning development, e.g., defining special kinds of output layers. If you want to work with existing layers, it may be better to use a higher-level library such as [Keras: https://keras.io].

Tensorflow is still new (1.0 came out in February 2017) and can be a bit of a moving target (particularly the contrib parts), but it includes some great tools such as Tensorboard for visualisation and is good at computing things on GPUs.

    1. Installing: ##

1. Install python of choice 2. `pip install tensorflow` 3. ?? 4. $$!

    1. Resources: ##

Here’s some useful resources to get started:

      1. Courses ###
      1. Books ###


      1. Blogs ###
      1. Examples: ###
Personal tools
Front page