This shows you the differences between two versions of the page.
Next revision | Previous revision | ||
cs501r_f2016:lab10 [2016/10/25 02:20] wingated created |
cs501r_f2016:lab10 [2021/06/30 23:42] (current) |
||
---|---|---|---|
Line 7: | Line 7: | ||
---- | ---- | ||
====Deliverable:==== | ====Deliverable:==== | ||
- | |||
- | {{ :cs501r_f2016:lab7_gan_results.png?200|}} | ||
For this lab, you will need to implement the `char-rnn` model of Karpathy. | For this lab, you will need to implement the `char-rnn` model of Karpathy. | ||
Line 14: | Line 12: | ||
show samples from the model. | show samples from the model. | ||
- | You should turn in an iPython notebook containing your code, along | + | You should turn in your python code (not necessarily in a notebook, if you'd rather not), along |
- | with novel text samples showing that your code is working: first, from | + | with novel text samples showing that your code is working: first, after training on |
the provided "alma.txt" dataset, and second, samples from your | the provided "alma.txt" dataset, and second, samples from your | ||
network trained on a dataset of your choice. | network trained on a dataset of your choice. | ||
An example of my final samples are shown below (more detail in the | An example of my final samples are shown below (more detail in the | ||
- | final section of this writeup), after 150 passed through the data. | + | final section of this writeup), after 150 passes through the data. |
Please generate about 15 samples for each dataset. | Please generate about 15 samples for each dataset. | ||
Line 58: | Line 56: | ||
This lab will help you develop several new tensorflow skills, as well | This lab will help you develop several new tensorflow skills, as well | ||
as understand some best practices needed for building large models. | as understand some best practices needed for building large models. | ||
- | In addition, we'll be able to create networks that generate neat images! | + | In addition, we'll be able to create networks that generate neat text! |
There are two parts of this lab: first, wiring up a basic | There are two parts of this lab: first, wiring up a basic | ||
Line 66: | Line 64: | ||
To help you get started, you should download and look at the | To help you get started, you should download and look at the | ||
- | [[http://liftothers.org/byu/lab8_scaffold.zip|Lab 8 scaffold code]] | + | [[http://liftothers.org/byu/lab8_scaffold.tar.gz|Lab 8 scaffold code]] |
which contains a few extra goodies to help you: a TextLoader class, | which contains a few extra goodies to help you: a TextLoader class, | ||
and some simple text files that you can use for debugging, and a | and some simple text files that you can use for debugging, and a | ||
little script called `switch.sh` that switches between text files. | little script called `switch.sh` that switches between text files. | ||
+ | |||
+ | After unpacking the scaffold tarball, you should run switch.sh to select one of the provided training files: | ||
+ | |||
+ | <code> | ||
+ | ./switch.sh alma.txt | ||
+ | </code> | ||
---- | ---- |