14 Commits

Author SHA1 Message Date
ac5d52295f nucleus sampling 2019-08-26 21:20:33 -07:00
f35fa1d920 push 774M model 2019-08-20 09:10:36 -07:00
c0859d7523 Fix TODO in sample.sample_sequences- Avoid 'leaving last token calculation to while loop' (#119)
* do initial run on full context

* decrement while loop iterations

* add context to output

* remove first param

* removing first param: change shape invariant
2019-05-30 21:49:18 -07:00
e5c5054474 allow models to be in a separate folder via models_dir argument (#129)
* models_dir argument to allow models in a separate folder

* default value for models_dir to be same as before

* allow environment variables and user home in models_dir
2019-05-16 09:42:58 -07:00
9d1e704fec slight fix to batch size description 2019-02-26 23:05:11 -08:00
ed49f03778 Add documentation for help flags (#81)
add description for flags
2019-02-26 23:01:38 -08:00
946facf551 fixed seed arg to ensure reproducibility in conditional-samples model 2019-02-20 14:51:14 -08:00
2cf46d997d fixed unconditional sampling reproducibility issue 2019-02-20 14:50:52 -08:00
2d0b62225c Update encoder.py to work on windows
This fixes https://github.com/openai/gpt-2/issues/26

```  File "C:\Users\James Pollack\Desktop\gpt-2\src\encoder.py", line 112, in get_encoder
    bpe_data = f.read()
  File "C:\Anaconda\envs\gpt-2\lib\encodings\cp1252.py", line 23, in decode
    return codecs.charmap_decode(input,self.errors,decoding_table)[0]
UnicodeDecodeError: 'charmap' codec can't decode byte 0x81 in position 66951: character maps to <undefined>```
2019-02-15 11:55:23 -08:00
77c7ae5ee5 Fix ModuleNotFoundError: No module named 'src' in the interactive_conditional_samples.py 2019-02-14 13:20:24 -08:00
bb617c436e Fix ModuleNotFoundError: No module named 'src' 2019-02-14 13:20:24 -08:00
7cdac144c3 fix bug and remove f strings 2019-02-14 11:34:14 -08:00
e33295b4b5 interact script for conditional samples 2019-02-14 10:40:41 -08:00
c2dae27c10 First commit 2019-02-10 23:47:49 -08:00