rhaymison commited on
Commit
be15288
1 Parent(s): 918e05d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +43 -3
README.md CHANGED
@@ -8,15 +8,24 @@ metrics:
8
  model-index:
9
  - name: flan-t5-base
10
  results: []
 
 
 
 
 
 
11
  ---
12
 
13
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
  should probably proofread and complete it, then remove this comment. -->
15
 
16
- # flan-t5-base
 
 
 
 
 
17
 
18
- This model is a fine-tuned version of [google/flan-t5-small](https://huggingface.co/google/flan-t5-small) on an unknown dataset.
19
- It achieves the following results on the evaluation set:
20
  - Loss: 1.6541
21
  - Rouge1: 16.3352
22
  - Rouge2: 6.2366
@@ -24,6 +33,22 @@ It achieves the following results on the evaluation set:
24
  - Rougelsum: 15.2755
25
  - Gen Len: 19.0
26
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
27
  ## Model description
28
 
29
  More information needed
@@ -81,3 +106,18 @@ The following hyperparameters were used during training:
81
  - Pytorch 2.2.1+cu121
82
  - Datasets 2.18.0
83
  - Tokenizers 0.15.2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8
  model-index:
9
  - name: flan-t5-base
10
  results: []
11
+ datasets:
12
+ - recogna-nlp/recognasumm
13
+ language:
14
+ - pt
15
+ library_name: transformers
16
+ pipeline_tag: summarization
17
  ---
18
 
19
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
20
  should probably proofread and complete it, then remove this comment. -->
21
 
22
+ # flan-t5-portuguese-small
23
+
24
+ This model aims to help supply the needs of models in the Portuguese language for certain tasks. The model presents a good performance for summary tasks.
25
+ Some errors due to word accentuation may occasionally occur due to the small version of the model.
26
+
27
+ model_max_length = 512
28
 
 
 
29
  - Loss: 1.6541
30
  - Rouge1: 16.3352
31
  - Rouge2: 6.2366
 
33
  - Rougelsum: 15.2755
34
  - Gen Len: 19.0
35
 
36
+ GPU: RTX 3060, 12GB, =~3500 cuda cores
37
+
38
+ ```python
39
+ !pip install transformers
40
+ from transformers import pipeline
41
+ summarization = pipeline("summarization", model="rhaymison/flan-t5-portuguese-summarization", tokenizer="rhaymison/flan-t5-portuguese-summarization")
42
+
43
+ prompt =f"""
44
+ sumarize: Na segunda disputa estadual para escolher o candidato do partido republicano para as eleições de novembro nos Estados Unidos, o ex-presidente Donald Trump teve mais uma vitória. Ele venceu as primárias em New Hampshire, que ocorreram na terça-feira (23/01). Antes disso, o favoritismo de Trump, apontado por diversas pesquisas, foi visto em sua vitória na primeira prévia do calendário eleitoral americano, em Iowa. Naquele Estado, Trump registrou 51% dos votos e vantagem de 30 pontos sobre o segundo colocado, o governador da Flórida, Ron DeSantis. No domingo (21/1), DeSantis anunciou sua desistência da corrida presidencial de 2024 e manifestou apoio a Trump. O movimento deixou Nikki Haley, ex-embaixadora dos Estados Unidos nas Nações Unidas, como a única rival significativa de Trump no partido.
45
+ """
46
+ output = summarization(prompt)
47
+
48
+ #Trump vence as primárias em New Hampshire, que ocorreram na terça-feira (23/01).
49
+ #Antes disso, o favoritismo de Trump, apontado por diversas pesquisas, foi visto ao segundo colocado, governador da Flórida, Ron DeSantis.
50
+ ```
51
+
52
  ## Model description
53
 
54
  More information needed
 
106
  - Pytorch 2.2.1+cu121
107
  - Datasets 2.18.0
108
  - Tokenizers 0.15.2
109
+
110
+ ### Comments
111
+
112
+ Any idea, help or report will always be welcome.
113
+
114
+ email: rhaymisoncristian@gmail.com
115
+
116
+ <div style="display:flex; flex-direction:row; justify-content:left">
117
+ <a href="https://www.linkedin.com/in/heleno-betini-2b3016175/" target="_blank">
118
+ <img src="https://img.shields.io/badge/LinkedIn-0077B5?style=for-the-badge&logo=linkedin&logoColor=white">
119
+ </a>
120
+ <a href="https://github.com/rhaymisonbetini" target="_blank">
121
+ <img src="https://img.shields.io/badge/GitHub-100000?style=for-the-badge&logo=github&logoColor=white">
122
+ </a>
123
+ </div>