This model is trained on roberta-base - https://huggingface.co/roberta-base

Dataset - https://drive.google.com/file/d/1hyXTTubD9CRjL1MBSIU_iVxFCdtGXqgB/view?usp=sharing

Notebook - https://colab.research.google.com/drive/1zHrs3hosTXBPiy0P1O-x6Z2EQ9RZGUsQ?usp=sharing

Label Precision Recall F1-Score Support
commodity 0.41 0.27 0.32 89
company 0.57 0.66 0.61 192
delivery_cap 0.00 0.00 0.00 10
delivery_location 0.57 0.17 0.27 46
delivery_port 0.66 0.85 0.74 608
delivery_state 0.85 0.48 0.62 120
incoterms 0.82 0.94 0.88 212
measures 0.75 0.77 0.76 682
package_type 0.84 0.90 0.87 338
pickup_cap 0.86 0.88 0.87 133
pickup_location 0.65 0.68 0.67 505
pickup_port 0.00 0.00 0.00 6
pickup_state 0.79 0.88 0.83 92
quantity 0.87 0.86 0.87 195
stackable 0.47 0.38 0.42 73
volume 0.36 0.31 0.33 71
weight 0.41 0.52 0.45 283
micro avg 0.69 0.73 0.71 3655
macro avg 0.58 0.56 0.56 3655
weighted avg 0.69 0.73 0.70 3655