update model card README.md

README.md

@@ -1,6 +1,5 @@
 ---
-license: apache-2.0
-base_model: distilbert-base-uncased
+base_model: allenai/scibert_scivocab_uncased
 tags:
 - generated_from_trainer
 metrics:
@@ -18,13 +17,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 # patentClassfication2
 
-This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
+This model is a fine-tuned version of [allenai/scibert_scivocab_uncased](https://huggingface.co/allenai/scibert_scivocab_uncased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6329
-- Accuracy: 0.6441
-- F1: 0.6528
-- Precision: 0.6402
-- Recall: 0.6658
+- Loss: 0.6359
+- Accuracy: 0.655
+- F1: 0.6955
+- Precision: 0.6304
+- Recall: 0.7756
 
 ## Model description
 
@@ -43,30 +42,22 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 5.310370197342976e-05
-- train_batch_size: 16
-- eval_batch_size: 16
+- learning_rate: 1.939963376695812e-05
+- train_batch_size: 8
+- eval_batch_size: 8
 - seed: 40
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 11
+- num_epochs: 3
 
 ### Training results
 
-| Training Loss | Epoch | Step  | Validation Loss | Accuracy | F1     | Precision | Recall |
-|:-------------:|:-----:|:-----:|:---------------:|:--------:|:------:|:---------:|:------:|
-| 0.652         | 1.0   | 2221  | 0.6470          | 0.6087   | 0.5218 | 0.6759    | 0.4249 |
-| 0.6283        | 2.0   | 4442  | 0.6329          | 0.6441   | 0.6528 | 0.6402    | 0.6658 |
-| 0.5689        | 3.0   | 6663  | 0.6334          | 0.6371   | 0.6788 | 0.6112    | 0.7631 |
-| 0.4785        | 4.0   | 8884  | 0.8592          | 0.6256   | 0.6131 | 0.6376    | 0.5905 |
-| 0.3792        | 5.0   | 11105 | 0.9728          | 0.6207   | 0.6098 | 0.6310    | 0.5901 |
-| 0.2839        | 6.0   | 13326 | 1.0226          | 0.6083   | 0.6212 | 0.6041    | 0.6394 |
-| 0.2211        | 7.0   | 15547 | 1.6336          | 0.6145   | 0.6067 | 0.6223    | 0.5919 |
-| 0.1756        | 8.0   | 17768 | 1.8340          | 0.6052   | 0.5951 | 0.6139    | 0.5773 |
-| 0.1451        | 9.0   | 19989 | 2.0495          | 0.6078   | 0.5980 | 0.6165    | 0.5807 |
-| 0.1147        | 10.0  | 22210 | 2.3889          | 0.6128   | 0.6128 | 0.6158    | 0.6098 |
-| 0.0983        | 11.0  | 24431 | 2.4921          | 0.6119   | 0.6132 | 0.6141    | 0.6123 |
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     | Precision | Recall |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
+| 0.6809        | 1.0   | 500  | 0.6729          | 0.625    | 0.5541 | 0.6997    | 0.4587 |
+| 0.6004        | 2.0   | 1000 | 0.6359          | 0.655    | 0.6955 | 0.6304    | 0.7756 |
+| 0.4696        | 3.0   | 1500 | 0.6658          | 0.675    | 0.6919 | 0.6673    | 0.7185 |
 
 
 ### Framework versions