liamcripwell commited on
Commit
cae4db1
1 Parent(s): 3748f59

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +48 -0
README.md ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - en
4
+ ---
5
+
6
+ # O->LED_para document simplification system
7
+
8
+ This is a pretrained version of the document simplification model presented in the Findings of ACL 2023 paper ["Context-Aware Document Simplification"](https://arxiv.org/abs/2305.06274).
9
+
10
+ It is a system based on the [Longformer encoder-decoder](https://huggingface.co/allenai/led-base-16384) that operates at the paragraph-level and is intended to be guided by a planner.
11
+
12
+ Target reading levels (1-4) should be indicated via a control token prepended to each input sequence ("\<RL_1\>", "\<RL_2\>", "\<RL_3\>", "\<RL_4\>"). If using the terminal interface, this will be handled automatically.
13
+
14
+ ## How to use
15
+ It is recommended to use the [plan_simp](https://github.com/liamcripwell/plan_simp/tree/main) library to interface with the model.
16
+
17
+ Here is how to load this model in PyTorch:
18
+
19
+ ```python
20
+ from plan_simp.models.bart import load_simplifier
21
+
22
+ simplifier, tokenizer, hparams = load_simplifier("liamcripwell/o-ledpara")
23
+ ```
24
+
25
+ Plan-guided generation and evaluation can be run from the terminal (see the repo for more details).
26
+
27
+ ```bash
28
+ python doc_simp/scripts/generate.py dynamic
29
+ --clf_model_ckpt=liamcripwell/pgdyn-plan
30
+ --model_ckpt=liamcripwell/o-ledpara
31
+ --test_file=<test_data>
32
+ --doc_id_col=pair_id
33
+ --context_dir=<context_dir>
34
+ --out_file=<output_csv>
35
+ --reading_lvl=s_level
36
+ --context_doc_id=pair_id
37
+ --para_lvl=True
38
+
39
+ python plan_simp/scripts/eval_simp.py
40
+ --input_data=newselaauto_docs_test.csv
41
+ --output_data=test_out_oledpara.csv
42
+ --x_col=complex_str
43
+ --r_col=simple_str
44
+ --y_col=pred
45
+ --doc_id_col=pair_id
46
+ --prepro=True
47
+ --sent_level=True
48
+ ```