itdainb commited on
Commit
d324fd3
·
verified ·
1 Parent(s): 05189a4

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -7
README.md CHANGED
@@ -29,14 +29,14 @@ pipeline_tag: text-classification
29
  ---
30
 
31
  #### Table of contents
32
- 1. [Installation](#install)
33
  2. [Pre-processing](#preprocess)
34
  3. [Usage with `sentence-transformers`](#sentence)
35
  4. [Usage with `transformers`](#transformers)
36
- 5. [Performance](#performance)
37
 
38
 
39
- ## Installation<a name="install"></a>
40
  - Install `pyvi` to word segment:
41
 
42
  - `pip install pyvi`
@@ -49,7 +49,7 @@ pipeline_tag: text-classification
49
 
50
  - `pip install transformers`
51
 
52
- ## Pre-processing<a name="preprocess"></a>
53
 
54
  ```python
55
  from pyvi import ViTokenizer
@@ -67,7 +67,7 @@ tokenized_sentences = [ViTokenizer.tokenize(sent) for sent in sentences]
67
  tokenized_pairs = [[tokenized_query, sent] for sent in tokenized_sentences]
68
  ```
69
 
70
- ## Usage with sentence-transformers<a name="sentence"></a>
71
 
72
  ```python
73
  from sentence_transformers import CrossEncoder
@@ -75,7 +75,7 @@ model = CrossEncoder('itdainb/vietnamese-cross-encoder', max_length=256)
75
  scores = model.predict(tokenized_pairs)
76
  ```
77
 
78
- ## Usage with transformers<a name="transformers"></a>
79
 
80
  ```python
81
  from transformers import AutoTokenizer, AutoModelForSequenceClassification
@@ -94,7 +94,7 @@ with torch.no_grad():
94
  print(scores)
95
  ```
96
 
97
- ## Performance<a name="performance"></a>
98
  In the following table, we provide various pre-trained Cross-Encoders together with their performance on the [MS MMarco Passage Reranking - Vi - Dev](https://huggingface.co/datasets/unicamp-dl/mmarco) dataset.
99
 
100
  | Model-Name | NDCG@3 | MRR@3 | NDCG@5 | MRR@5 | NDCG@10 | MRR@10 | Docs / Sec |
 
29
  ---
30
 
31
  #### Table of contents
32
+ 1. [Installation](#Installation)
33
  2. [Pre-processing](#preprocess)
34
  3. [Usage with `sentence-transformers`](#sentence)
35
  4. [Usage with `transformers`](#transformers)
36
+ 5. [Performance](#Performance)
37
 
38
 
39
+ ## Installation
40
  - Install `pyvi` to word segment:
41
 
42
  - `pip install pyvi`
 
49
 
50
  - `pip install transformers`
51
 
52
+ ## Pre-processing
53
 
54
  ```python
55
  from pyvi import ViTokenizer
 
67
  tokenized_pairs = [[tokenized_query, sent] for sent in tokenized_sentences]
68
  ```
69
 
70
+ ## Usage with sentence-transformers
71
 
72
  ```python
73
  from sentence_transformers import CrossEncoder
 
75
  scores = model.predict(tokenized_pairs)
76
  ```
77
 
78
+ ## Usage with transformers
79
 
80
  ```python
81
  from transformers import AutoTokenizer, AutoModelForSequenceClassification
 
94
  print(scores)
95
  ```
96
 
97
+ ## Performance
98
  In the following table, we provide various pre-trained Cross-Encoders together with their performance on the [MS MMarco Passage Reranking - Vi - Dev](https://huggingface.co/datasets/unicamp-dl/mmarco) dataset.
99
 
100
  | Model-Name | NDCG@3 | MRR@3 | NDCG@5 | MRR@5 | NDCG@10 | MRR@10 | Docs / Sec |