Upload 5 files

Tamil 11-Class Emotion Classifier

A fine-tuned Tamil language model that detects 11 emotions from text.

Supported emotions:
- Ambiguous
- Anger
- Anticipation
- Disgust
- Fear
- Joy
- Love
- Neutral
- Sadness
- Surprise
- Trust

Performance (validation set):
→ Accuracy: 94.5%
→ Macro F1: ~94.3%
→ Best checkpoint: epoch 6

Base model: jusgowiturs/autotrain-tamil_emotion_11_tamilbert-2710380899

Trained on ~53k balanced Tamil examples (6 epochs, lr 2e-5, batch 16)

Best suited for:
• Tamil chatbots with emotion understanding
• Social media sentiment analysis in Tamil
• Mental health & customer support tools
• Tamil NLP research & education

Limitations:
• Works best on short–medium length sentences
• Code-mixed Tamil–English may reduce accuracy
• Sarcasm, irony and heavy dialect variations remain challenging

Open for testing, further fine-tuning and community improvements.

Made with ❤️ for the Tamil NLP community
— Bimsara Serasinghe, March 2026

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

config.json

@@ -0,0 +1,59 @@
+{
+  "_num_labels": 11,
+  "add_cross_attention": false,
+  "architectures": [
+    "XLMRobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "dtype": "float32",
+  "eos_token_id": 2,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "Ambiguous",
+    "1": "Anger",
+    "2": "Anticipation",
+    "3": "Disguist",
+    "4": "Fear",
+    "5": "Joy",
+    "6": "Love",
+    "7": "Neutral",
+    "8": "Sadness",
+    "9": "Surprise",
+    "10": "Trust"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "is_decoder": false,
+  "label2id": {
+    "Ambiguous": 0,
+    "Anger": 1,
+    "Anticipation": 2,
+    "Disguist": 3,
+    "Fear": 4,
+    "Joy": 5,
+    "Love": 6,
+    "Neutral": 7,
+    "Sadness": 8,
+    "Surprise": 9,
+    "Trust": 10
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "xlm-roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "padding": "max_length",
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "tie_word_embeddings": true,
+  "transformers_version": "5.1.0",
+  "type_vocab_size": 1,
+  "use_cache": false,
+  "vocab_size": 250002
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0121578e971627f881b7e1ebd571724273f154fc1b654d7097e617a7f2c7bdfd
+size 1112232668

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0a8f22c7c06d9a64d49b4a0571324ba43d4da6dc988ff38daf31d33586fc5f1
+size 16766399

tokenizer_config.json

@@ -0,0 +1,14 @@
+{
+  "add_prefix_space": true,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "is_local": false,
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d626af054f81add74b6c4193013b8e178af26dd5a09cb2880c6e9c7c0125e8d
+size 5201