AngelinaZanardi/educational_value_fasttext_autonomous_nno

Educational Score FastText Model

• Trained on AngelinaZanardi/fineweb-kimi-k2-instruct-nno_cleaned
• Target column: educational_score
• Hyperparameters: lr=0.05, epoch=50, wordNgrams=1, dim=300, minCount=5, loss=softmax, ws=7, minn=3, maxn=6
• Validation F1: 0.4976
• Test F1: 0.4938

✅ Validation Weighted F1: 0.4976 ✅ Test Weighted F1: 0.4938

Confusion Matrix: [[ 3 15 0 0 0 0] [ 3 171 4 24 0 0] [ 0 54 7 51 4 0] [ 0 28 11 115 8 0] [ 0 2 0 28 10 0] [ 0 0 0 6 5 0]]

Classification Report: precision recall f1-score support

       0       0.50      0.17      0.25        18
       1       0.63      0.85      0.72       202
       2       0.32      0.06      0.10       116
       3       0.51      0.71      0.60       162
       4       0.37      0.25      0.30        40
       5       0.00      0.00      0.00        11

accuracy                           0.56       549

macro avg 0.39 0.34 0.33 549 weighted avg 0.50 0.56 0.49 549