Update README.md

README.md

@@ -1,17 +1,23 @@
 ---
 pipeline_tag: sentence-similarity
+language: 
+  - 'pt'
 tags:
 - sentence-transformers
 - feature-extraction
 - sentence-similarity
 - transformers
+datasets:
+- 'unicamp-dl/mmarco'
 
 ---
 
-# {MODEL_NAME}
+# mpjan/msmarco-distilbert-base-tas-b-mmarco-pt-300k
 
 This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.
 
+It is a fine-tuning of [sentence-transformers/msmarco-distilbert-base-tas-b](https://huggingface.co/sentence-transformers/msmarco-distilbert-base-tas-b) on the first 300k triplets of the Portuguese subset in [unicamp-dl/mmarco](https://huggingface.co/datasets/unicamp-dl/mmarco).
+
 <!--- Describe your model here -->
 
 ## Usage (Sentence-Transformers)
@@ -28,7 +34,7 @@ Then you can use the model like this:
 from sentence_transformers import SentenceTransformer
 sentences = ["This is an example sentence", "Each sentence is converted"]
 
-model = SentenceTransformer('{MODEL_NAME}')
+model = SentenceTransformer('mpjan/msmarco-distilbert-base-tas-b-mmarco-pt-300k')
 embeddings = model.encode(sentences)
 print(embeddings)
 ```
@@ -51,8 +57,8 @@ def cls_pooling(model_output, attention_mask):
 sentences = ['This is an example sentence', 'Each sentence is converted']
 
 # Load model from HuggingFace Hub
-tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}')
-model = AutoModel.from_pretrained('{MODEL_NAME}')
+tokenizer = AutoTokenizer.from_pretrained('mpjan/msmarco-distilbert-base-tas-b-mmarco-pt-300k')
+model = AutoModel.from_pretrained('mpjan/msmarco-distilbert-base-tas-b-mmarco-pt-300k')
 
 # Tokenize sentences
 encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')