Lautaro commited on
Commit
e54bf0f
1 Parent(s): a4c2c8d

Update Readme

Browse files
Files changed (1) hide show
  1. README.md +9 -12
README.md CHANGED
@@ -13,10 +13,13 @@ widget:
13
  ---
14
 
15
 
16
- # A zero-shot classifier based on bertin-roberta-base-finetuning-esnli
 
 
 
17
 
18
  ## Usage (HuggingFace Transformers)
19
- Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.
20
 
21
  ```python
22
  from transformers import pipeline
@@ -26,7 +29,7 @@ classifier = pipeline("zero-shot-classification",
26
  classifier(
27
  "El autor se perfila, a los 50 años de su muerte, como uno de los grandes de su siglo",
28
  candidate_labels=["cultura", "sociedad", "economia", "salud", "deportes"],
29
- hypothesis_template="Este ejemplo es {}."
30
  )
31
  ```
32
 
@@ -34,6 +37,8 @@ The `hypothesis_template` parameter is important and should be in Spanish. **In
34
 
35
  ## Training
36
 
 
 
37
  **Dataset**
38
 
39
  We used a collection of datasets of Natural Language Inference as training data:
@@ -41,15 +46,7 @@ We used a collection of datasets of Natural Language Inference as training data:
41
  - [SNLI](https://nlp.stanford.edu/projects/snli/), automatically translated
42
  - [MultiNLI](https://cims.nyu.edu/~sbowman/multinli/), automatically translated
43
 
44
- The whole dataset used is available [here](https://huggingface.co/datasets/hackathon-pln-es/ESnli).
45
-
46
- ## Full Model Architecture
47
- ```
48
- SentenceTransformer(
49
- (0): Transformer({'max_seq_length': 514, 'do_lower_case': False}) with Transformer model: RobertaModel
50
- (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
51
- )
52
- ```
53
 
54
  ## Authors
55
 
 
13
  ---
14
 
15
 
16
+ # A zero-shot classifier based on bertin-roberta-base-spanish
17
+ This model was trained on the basis of the model `bertin-roberta-base-spanish` using **Cross encoder** for NLI task. A CrossEncoder takes a sentence pair as input and outputs a label so it learns to predict the labels: "contradiction": 0, "entailment": 1, "neutral": 2.
18
+
19
+ You can use it with Hugging Face's Zero-shot pipeline to make **zero-shot classifications**. Given a sentence and an arbitrary set of labels/topics, it will output the likelihood of the sentence belonging to each of the topic.
20
 
21
  ## Usage (HuggingFace Transformers)
22
+ The simplest way to use the model is the huggingface transformers pipeline tool. Just initialize the pipeline specifying the task as "zero-shot-classification" and select "hackathon-pln-es/bertin-roberta-base-zeroshot-esnli" as model.
23
 
24
  ```python
25
  from transformers import pipeline
 
29
  classifier(
30
  "El autor se perfila, a los 50 años de su muerte, como uno de los grandes de su siglo",
31
  candidate_labels=["cultura", "sociedad", "economia", "salud", "deportes"],
32
+ hypothesis_template="Esta oración es sobre {}."
33
  )
34
  ```
35
 
 
37
 
38
  ## Training
39
 
40
+ We used [sentence-transformers](https://www.SBERT.net) to train the model.
41
+
42
  **Dataset**
43
 
44
  We used a collection of datasets of Natural Language Inference as training data:
 
46
  - [SNLI](https://nlp.stanford.edu/projects/snli/), automatically translated
47
  - [MultiNLI](https://cims.nyu.edu/~sbowman/multinli/), automatically translated
48
 
49
+ The whole dataset used is available [here](https://huggingface.co/datasets/hackathon-pln-es/nli-es).
 
 
 
 
 
 
 
 
50
 
51
  ## Authors
52