multilingual upgrade upload of sarcasm-detector

README.md

@@ -1,16 +1,18 @@
 ---
-language: "en"
+language: "multilingual"
 tags:
 - bert
 - sarcasm-detection
 - text-classification
 widget:
+- text: "Gli Usa a un passo dalla recessione"
 - text: "CIA Realizes It's Been Using Black Highlighters All These Years."
+- text: "We deden een man een nacht in een vat met cola en nu is hij dood"
 ---
 
-# English Sarcasm Detector
+# Multilingual Sarcasm Detector
 
-English Sarcasm Detector is a text classification model built to detect sarcasm from news article titles. It is fine-tuned on [bert-base-uncased](https://huggingface.co/bert-base-uncased) and the training data consists of ready-made dataset available on Kaggle.
+Multilingual Sarcasm Detector is a text classification model built to detect sarcasm from news article titles. It is fine-tuned on [bert-base-multilingual-uncased](https://huggingface.co/bert-base-multilingual-uncased) and the training data consists of ready-made datasets available on Kaggle as well scraped data from multiple newspapers in English, Dutch and Italian.
 
 
 <b>Labels</b>: 
@@ -19,15 +21,22 @@ English Sarcasm Detector is a text classification model built to detect sarcasm
 
 
 ## Source Data
-
+	  
 Datasets:
 - English language data: [Kaggle: News Headlines Dataset For Sarcasm Detection](https://www.kaggle.com/datasets/rmisra/news-headlines-dataset-for-sarcasm-detection).
+- Dutch non-sarcastic data: [Kaggle: Dutch News Articles](https://www.kaggle.com/datasets/maxscheijen/dutch-news-articles)
+
+Scraped data:
+- Dutch sarcastic news from [De Speld](https://speld.nl)
+- Italian non-sarcastic news from [Il Giornale](https://www.ilgiornale.it)
+- Italian sarcastic news from [Lercio](https://www.lercio.it)
 
 ## Training Dataset
 - [helinivan/sarcasm_headlines_multilingual](https://huggingface.co/datasets/helinivan/sarcasm_headlines_multilingual)
 
 ## Codebase:
-- Git Repo: [Official repository](https://github.com/helinivan/multilingual-sarcasm-detector).
+- Git Repo: [Official repository](https://github.com/helinivan/multilingual-sarcasm-detector)
+
 
 ---
 
@@ -41,7 +50,7 @@ import string
 def preprocess_data(text: str) -> str:
    return text.lower().translate(str.maketrans("", "", string.punctuation)).strip()
 
-MODEL_PATH = "helinivan/english-sarcasm-detector"
+MODEL_PATH = "helinivan/multilingual-sarcasm-detector"
 
 tokenizer = AutoTokenizer.from_pretrained(MODEL_PATH)
 model = AutoModelForSequenceClassification.from_pretrained(MODEL_PATH)
@@ -59,13 +68,13 @@ results = {"is_sarcastic": prediction, "confidence": confidence}
 Output: 
 
 ```
-{'is_sarcastic': 1, 'confidence': 0.9337034225463867}
+{'is_sarcastic': 1, 'confidence': 0.9374828934669495}
 ```
 
 ## Performance
 | Model-Name | F1 | Precision | Recall | Accuracy 
 | ------------- |:-------------| -----| -----| ----| 
-| [helinivan/english-sarcasm-detector ](https://huggingface.co/helinivan/english-sarcasm-detector)| **92.38** | 92.75 | 92.38 | 92.42
+| [helinivan/english-sarcasm-detector ](https://huggingface.co/helinivan/english-sarcasm-detector)| 92.38 | 92.75 | 92.38 | 92.42
 | [helinivan/italian-sarcasm-detector ](https://huggingface.co/helinivan/italian-sarcasm-detector) | 88.26 | 87.66 | 89.66 | 88.69
-| [helinivan/multilingual-sarcasm-detector ](https://huggingface.co/helinivan/multilingual-sarcasm-detector) | 87.23 | 88.65 | 86.33 | 88.30
+| [helinivan/multilingual-sarcasm-detector ](https://huggingface.co/helinivan/multilingual-sarcasm-detector) | **87.23** | 88.65 | 86.33 | 88.30
 | [helinivan/dutch-sarcasm-detector ](https://huggingface.co/helinivan/dutch-sarcasm-detector) | 83.02 | 84.27 | 82.01 | 86.81

config.json

@@ -1,11 +1,11 @@
 {
-  "_name_or_path": "bert-base-uncased",
+  "_name_or_path": "bert-base-multilingual-uncased",
   "architectures": [
     "BertForSequenceClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
   "classifier_dropout": null,
-  "gradient_checkpointing": false,
+  "directionality": "bidi",
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
@@ -17,11 +17,16 @@
   "num_attention_heads": 12,
   "num_hidden_layers": 12,
   "pad_token_id": 0,
+  "pooler_fc_size": 768,
+  "pooler_num_attention_heads": 12,
+  "pooler_num_fc_layers": 3,
+  "pooler_size_per_head": 128,
+  "pooler_type": "first_token_transform",
   "position_embedding_type": "absolute",
   "problem_type": "single_label_classification",
   "torch_dtype": "float32",
   "transformers_version": "4.24.0",
   "type_vocab_size": 2,
   "use_cache": true,
-  "vocab_size": 30522
+  "vocab_size": 105879
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:93827e574975e434a95b6b5b9c9ffbae9132f863a0dc467b89121e88f72f1fe2
-size 438006125
+oid sha256:27bbe03121cad8bd4234e90b72aa504dd44fa2d0fd993ced22cba5208cab33ca
+size 669502829

tokenizer_config.json

@@ -4,7 +4,7 @@
   "do_lower_case": true,
   "mask_token": "[MASK]",
   "model_max_length": 512,
-  "name_or_path": "bert-base-uncased",
+  "name_or_path": "bert-base-multilingual-uncased",
   "never_split": null,
   "pad_token": "[PAD]",
   "sep_token": "[SEP]",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f964b93090dcfc704c2bc0c2a87b6992d5ef9f462e37dd6e4e2969c383cf0284
-size 3311
+oid sha256:8fd7b1893ae4abf4769058ee62f97442a4fffade3dbe564e7b76c86904c601da
+size 3375