Mainak Manna
commited on
Commit
•
e6d9b25
1
Parent(s):
5a9d376
First version of the model
Browse files
README.md
CHANGED
@@ -6,7 +6,7 @@ tags:
|
|
6 |
datasets:
|
7 |
- jrc-acquis
|
8 |
widget:
|
9 |
-
- text: "
|
10 |
|
11 |
---
|
12 |
|
@@ -38,7 +38,7 @@ tokenizer=AutoTokenizer.from_pretrained(pretrained_model_name_or_path = "SEBIS/l
|
|
38 |
device=0
|
39 |
)
|
40 |
|
41 |
-
en_text = "
|
42 |
|
43 |
pipeline([en_text], max_length=512)
|
44 |
```
|
@@ -48,11 +48,13 @@ pipeline([en_text], max_length=512)
|
|
48 |
The legal_t5_small_summ_en model was trained on [JRC-ACQUIS](https://wt-public.emm4u.eu/Acquis/index_2.2.html) dataset consisting of 22 Thousand texts.
|
49 |
|
50 |
## Training procedure
|
|
|
51 |
|
|
|
52 |
### Preprocessing
|
53 |
|
54 |
### Pretraining
|
55 |
-
|
56 |
|
57 |
|
58 |
## Evaluation results
|
@@ -67,3 +69,5 @@ Test results :
|
|
67 |
|
68 |
|
69 |
### BibTeX entry and citation info
|
|
|
|
|
|
6 |
datasets:
|
7 |
- jrc-acquis
|
8 |
widget:
|
9 |
+
- text: "on the illegal trade in firearms (SCH/Com-ex (99) 10) THE EXECUTIVE COMMITTEE, Having regard to Article 132 of the Convention implementing the Schengen Agreement, Having regard to Article 9 of the abovementioned Convention, HAS DECIDED AS FOLLOWS: Henceforth, the Contracting Parties shall submit each year by 31 July their national annual data for the preceding year on illegal trade in firearms, on the basis of the joint table for compiling statistics annexed to document SCH/I-ar (98) 32. Luxembourg, 28 April 1999. The Chairman C. H. Schapper ANNEX SCH/I-Ar (98) 32 %gt%PIC FILE= %quot%L_2000239EN.047002.EPS%quot%%gt% %gt%PIC FILE= %quot%L_2000239EN.047101.EPS%quot%%gt% %gt%PIC FILE= %quot%L_2000239EN.047201.EPS%quot%%gt% %gt%PIC FILE= %quot%L_2000239EN.047301.EPS%quot%%gt% "
|
10 |
|
11 |
---
|
12 |
|
|
|
38 |
device=0
|
39 |
)
|
40 |
|
41 |
+
en_text = "on the illegal trade in firearms (SCH/Com-ex (99) 10) THE EXECUTIVE COMMITTEE, Having regard to Article 132 of the Convention implementing the Schengen Agreement, Having regard to Article 9 of the abovementioned Convention, HAS DECIDED AS FOLLOWS: Henceforth, the Contracting Parties shall submit each year by 31 July their national annual data for the preceding year on illegal trade in firearms, on the basis of the joint table for compiling statistics annexed to document SCH/I-ar (98) 32. Luxembourg, 28 April 1999. The Chairman C. H. Schapper ANNEX SCH/I-Ar (98) 32 %gt%PIC FILE= %quot%L_2000239EN.047002.EPS%quot%%gt% %gt%PIC FILE= %quot%L_2000239EN.047101.EPS%quot%%gt% %gt%PIC FILE= %quot%L_2000239EN.047201.EPS%quot%%gt% %gt%PIC FILE= %quot%L_2000239EN.047301.EPS%quot%%gt% "
|
42 |
|
43 |
pipeline([en_text], max_length=512)
|
44 |
```
|
|
|
48 |
The legal_t5_small_summ_en model was trained on [JRC-ACQUIS](https://wt-public.emm4u.eu/Acquis/index_2.2.html) dataset consisting of 22 Thousand texts.
|
49 |
|
50 |
## Training procedure
|
51 |
+
An unigram model trained with 88M lines of text from the parallel corpus (of all possible language pairs) to get the vocabulary (with byte pair encoding), which is used with this model.
|
52 |
|
53 |
+
The model was trained on a single TPU Pod V3-8 for 250K steps in total, using sequence length 512 (batch size 64). It has a total of approximately 220M parameters and was trained using the encoder-decoder architecture. The optimizer used is AdaFactor with inverse square root learning rate schedule for pre-training.
|
54 |
### Preprocessing
|
55 |
|
56 |
### Pretraining
|
57 |
+
|
58 |
|
59 |
|
60 |
## Evaluation results
|
|
|
69 |
|
70 |
|
71 |
### BibTeX entry and citation info
|
72 |
+
|
73 |
+
> Created by [Ahmed Elnaggar/@Elnaggar_AI](https://twitter.com/Elnaggar_AI) | [LinkedIn](https://www.linkedin.com/in/prof-ahmed-elnaggar/)
|