Training in progress epoch 4

README.md

@@ -15,17 +15,17 @@ 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 an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 0.0823
-- Train Accuracy: 0.9705
-- Train F1 M: 0.5511
-- Train Precision M: 0.4024
-- Train Recall M: 0.9370
-- Validation Loss: 0.2550
-- Validation Accuracy: 0.9116
-- Validation F1 M: 0.5567
-- Validation Precision M: 0.4057
-- Validation Recall M: 0.9330
-- Epoch: 3
+- Train Loss: 0.0532
+- Train Accuracy: 0.9812
+- Train F1 M: 0.5544
+- Train Precision M: 0.4027
+- Train Recall M: 0.9558
+- Validation Loss: 0.2580
+- Validation Accuracy: 0.9175
+- Validation F1 M: 0.5588
+- Validation Precision M: 0.4059
+- Validation Recall M: 0.9423
+- Epoch: 4
 
 ## Model description
 
@@ -55,6 +55,7 @@ The following hyperparameters were used during training:
 | 0.1974     | 0.9259         | 0.5184     | 0.3930            | 0.8161         | 0.1978          | 0.9202              | 0.5425          | 0.4014                 | 0.8778              | 1     |
 | 0.1242     | 0.9551         | 0.5382     | 0.3974            | 0.8918         | 0.1970          | 0.9248              | 0.5583          | 0.4106                 | 0.9195              | 2     |
 | 0.0823     | 0.9705         | 0.5511     | 0.4024            | 0.9370         | 0.2550          | 0.9116              | 0.5567          | 0.4057                 | 0.9330              | 3     |
+| 0.0532     | 0.9812         | 0.5544     | 0.4027            | 0.9558         | 0.2580          | 0.9175              | 0.5588          | 0.4059                 | 0.9423              | 4     |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:852138cbdf01dd301c9d31cee2fac0f71817471b4850bf14f56848e913303492
+oid sha256:b8dfc3104ba71c1cc15e0e168df0d132af17159e43fdaaf27035f96998936e7e
 size 438223128