update model card README.md

README.md

@@ -0,0 +1,66 @@
+---
+license: apache-2.0
+tags:
+- generated_from_trainer
+metrics:
+- f1
+- accuracy
+model-index:
+- name: distilbert-base-uncased_research_articles_multilabel
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# distilbert-base-uncased_research_articles_multilabel
+
+This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.1956
+- F1: 0.8395
+- Roc Auc: 0.8909
+- Accuracy: 0.6977
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 64
+- eval_batch_size: 64
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 5
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | F1     | Roc Auc | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:--------:|
+| 0.3043        | 1.0   | 263  | 0.2199          | 0.8198 | 0.8686  | 0.6829   |
+| 0.2037        | 2.0   | 526  | 0.1988          | 0.8355 | 0.8845  | 0.7010   |
+| 0.1756        | 3.0   | 789  | 0.1956          | 0.8395 | 0.8909  | 0.6977   |
+| 0.1579        | 4.0   | 1052 | 0.1964          | 0.8371 | 0.8902  | 0.6919   |
+| 0.1461        | 5.0   | 1315 | 0.1991          | 0.8353 | 0.8874  | 0.6953   |
+
+
+### Framework versions
+
+- Transformers 4.21.3
+- Pytorch 1.12.1
+- Datasets 2.4.0
+- Tokenizers 0.12.1