bert-1-epoch-sentiment

README.md

@@ -24,13 +24,13 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.7028655161185282
+      value: 0.6895962227287529
     - name: Precision
       type: precision
-      value: 0.7052777584989943
+      value: 0.6932981822495374
     - name: Recall
       type: recall
-      value: 0.7028655161185282
+      value: 0.6895962227287529
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -40,12 +40,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the tweet_eval dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6558
-- Accuracy: 0.7029
-- Precision: 0.7053
-- Recall: 0.7029
-- Micro-avg-recall: 0.7029
-- Micro-avg-precision: 0.7029
+- Loss: 0.6998
+- Accuracy: 0.6896
+- Precision: 0.6933
+- Recall: 0.6896
+- Micro-avg-recall: 0.6896
+- Micro-avg-precision: 0.6896
 
 ## Model description
 
@@ -76,7 +76,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | Micro-avg-recall | Micro-avg-precision |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:----------------:|:-------------------:|
-| 0.5503        | 1.0   | 2851 | 0.6558          | 0.7029   | 0.7053    | 0.7029 | 0.7029           | 0.7029              |
+| 0.5756        | 1.0   | 2851 | 0.6998          | 0.6896   | 0.6933    | 0.6896 | 0.6896           | 0.6896              |
 
 
 ### Framework versions