Hasil Train Train Loop

Epoch Train Loss Train ACC Train F1 Train REC Train PRE Valid Loss Valid ACC Valid F1 Valid REC Valid PRE Catatan
1 0.7815 0.6594 0.5856 0.5927 0.6117 0.6883 0.7052 0.6146 0.6285 0.7060 Model terbaik disimpan
2 0.5599 0.7702 0.7224 0.7199 0.7331 0.5264 0.7932 0.7493 0.7435 0.7728 Model terbaik disimpan
3 0.4428 0.8238 0.7877 0.7843 0.7942 0.4719 0.8140 0.7573 0.7519 0.8106 Model terbaik disimpan
4 0.3586 0.8560 0.8260 0.8224 0.8317 0.4138 0.8395 0.8114 0.8098 0.8155 Model terbaik disimpan
5 0.3066 0.8783 0.8532 0.8504 0.8568 0.3739 0.8542 0.8277 0.8253 0.8307 Model terbaik disimpan
6 0.2521 0.8997 0.8787 0.8766 0.8812 0.3728 0.8642 0.8439 0.8482 0.8417 Model terbaik disimpan
7 0.2203 0.9144 0.8956 0.8928 0.8989 0.3827 0.8650 0.8430 0.8442 0.8436 VALID LOSS tidak membaik (1/2)
8 0.1811 0.9323 0.9177 0.9158 0.9198 0.3579 0.8696 0.8450 0.8441 0.8461 Model terbaik disimpan
9 0.1684 0.9336 0.9189 0.9168 0.9213 0.3645 0.8773 0.8585 0.8620 0.8566 VALID LOSS tidak membaik (1/2)
10 0.1619 0.9384 0.9251 0.9235 0.9268 0.4715 0.8418 0.8158 0.8154 0.8195 VALID LOSS tidak membaik (2/2) — Early stopping

Accuracy per Epoch

image/png

Loss per Epoch

image/png

Classification Report pada Data Testing

image/png

Confusion Matrix pada Data Testing

image/png

Distribusi Sentimen Hasil Pred Pada Data Testing

image/png

WordCloud Hasil Prediksi Pada Data Testing

image/png

Frekuensi Kata Hasil Prediksi Pada Data Testing

image/png

Downloads last month
8
Safetensors
Model size
0.1B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Sahron/analisis-sentiment-indobert8696

Finetuned
(103)
this model