Update README.md

README.md

@@ -1,3 +1,32 @@
+# This model is a fine-tuned version of distilbert/distilbert-base-uncased for NER Task. 
+
+### The following hyperparameters were used during training:
+
+- learning_rate=2e-5,
+- per_device_train_batch_size=128,
+- per_device_eval_batch_size=128,
+- num_train_epochs=4,
+- weight_decay=0.01,
+- num_epochs: 4
+
+### It achieves the following results on the evaluation set:
+
+#### Training results
+
+| Epoch | Training Loss | Validation Loss | Precision |  Recall  |    F1    | Accuracy |
+|:-----:|:-------------:|:---------------:|:---------:|:--------:|:--------:|:--------:|
+|   1   |    0.286800   |     0.125389    |  0.977472 | 0.977472 | 0.977472 | 0.977472 |
+|   2   |    0.051800   |     0.026039    |  0.993742 | 0.993742 | 0.993742 | 0.993742 |
+|   3   |    0.024200   |     0.013263    |  0.994994 | 0.994994 | 0.994994 | 0.994994 |
+|   4   |    0.015500   |     0.013252    |  0.994994 | 0.994994 | 0.994994 | 0.994994 |
+
+
+Framework versions
+Transformers 4.41.2
+Pytorch 2.3.0+cu121
+Datasets 2.19.2
+Tokenizers 0.19.1
+
 ---
 title: Marketplace Products
 emoji: 🚀