pszemraj commited on
Commit
f8aa77a
1 Parent(s): 0a4e36b

add gist with custom class

Browse files
Files changed (1) hide show
  1. README.md +5 -4
README.md CHANGED
@@ -55,9 +55,13 @@ inference:
55
 
56
  This model is a fine-tuned version of [google/flan-t5-xl](https://huggingface.co/google/flan-t5-xl) on an extended version of the `JFLEG` dataset.
57
 
 
 
 
 
58
  ![ex](https://i.imgur.com/5QGGF0Z.png)
 
59
 
60
- > the API can be slow due to model size, try the notebook!
61
 
62
  ## Model description
63
 
@@ -71,9 +75,6 @@ Compare some of the more severe error examples on [other grammar correction mode
71
  - Model: `apache-2.0`
72
  - currently **a work in progress**! While probably useful for "single-shot grammar correction" in many cases, **check the output for correctness, ok?**.
73
 
74
- ## Training and evaluation data
75
-
76
- More information needed
77
 
78
  ## Training procedure
79
 
 
55
 
56
  This model is a fine-tuned version of [google/flan-t5-xl](https://huggingface.co/google/flan-t5-xl) on an extended version of the `JFLEG` dataset.
57
 
58
+ - [here is a custom class wrapper](https://gist.github.com/pszemraj/14f7b13bd2d953176db2371e5d320915) that makes using this with `bitsandbytes` easier
59
+ - the API can be slow due to model size, try [the notebook](https://colab.research.google.com/gist/pszemraj/43fc6a5c5acd94a3d064384dd1f3654c/demo-flan-t5-xl-grammar-synthesis.ipynb)!
60
+
61
+ <br>
62
  ![ex](https://i.imgur.com/5QGGF0Z.png)
63
+ <br>
64
 
 
65
 
66
  ## Model description
67
 
 
75
  - Model: `apache-2.0`
76
  - currently **a work in progress**! While probably useful for "single-shot grammar correction" in many cases, **check the output for correctness, ok?**.
77
 
 
 
 
78
 
79
  ## Training procedure
80