Model save

README.md

@@ -18,14 +18,14 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google-bert/bert-base-cased](https://huggingface.co/google-bert/bert-base-cased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6472
-- F1 Macro: 0.9202
-- F1: 0.9481
-- F1 Neg: 0.8923
-- Acc: 0.93
-- Prec: 0.9446
-- Recall: 0.9517
-- Mcc: 0.8405
+- Loss: 0.3688
+- F1 Macro: 0.9065
+- F1: 0.9385
+- F1 Neg: 0.8745
+- Acc: 0.9175
+- Prec: 0.9403
+- Recall: 0.9368
+- Mcc: 0.8131
 
 ## Model description
 
@@ -51,18 +51,20 @@ The following hyperparameters were used during training:
 - distributed_type: multi-GPU
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 5
+- num_epochs: 7
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1     | F1 Neg | Acc    | Prec   | Recall | Mcc    |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:------:|:------:|:------:|:------:|
-| 0.227         | 1.0   | 1647 | 0.4468          | 0.8956   | 0.9225 | 0.8687 | 0.9025 | 0.9393 | 0.9062 | 0.7922 |
-| 0.0771        | 2.0   | 3294 | 0.7307          | 0.8813   | 0.9216 | 0.8409 | 0.895  | 0.8821 | 0.9648 | 0.7706 |
-| 0.0192        | 3.0   | 4941 | 0.7794          | 0.8969   | 0.9258 | 0.8681 | 0.905  | 0.9258 | 0.9258 | 0.7938 |
-| 0.011         | 4.0   | 6588 | 0.9531          | 0.8893   | 0.9194 | 0.8591 | 0.8975 | 0.9249 | 0.9141 | 0.7787 |
-| 0.0028        | 5.0   | 8235 | 1.0420          | 0.8927   | 0.9206 | 0.8649 | 0.9    | 0.9355 | 0.9062 | 0.7863 |
+| Training Loss | Epoch | Step  | Validation Loss | F1 Macro | F1     | F1 Neg | Acc    | Prec   | Recall | Mcc    |
+|:-------------:|:-----:|:-----:|:---------------:|:--------:|:------:|:------:|:------:|:------:|:------:|:------:|
+| 0.2307        | 1.0   | 1647  | 0.3575          | 0.9168   | 0.9386 | 0.8949 | 0.9225 | 0.9518 | 0.9258 | 0.8342 |
+| 0.0661        | 2.0   | 3294  | 0.6221          | 0.8878   | 0.9248 | 0.8507 | 0.9    | 0.8913 | 0.9609 | 0.7811 |
+| 0.0334        | 3.0   | 4941  | 0.8234          | 0.8964   | 0.9215 | 0.8713 | 0.9025 | 0.9502 | 0.8945 | 0.7956 |
+| 0.0209        | 4.0   | 6588  | 0.7564          | 0.9008   | 0.9310 | 0.8705 | 0.91   | 0.9135 | 0.9492 | 0.8029 |
+| 0.0104        | 5.0   | 8235  | 0.8753          | 0.8984   | 0.9243 | 0.8725 | 0.905  | 0.9431 | 0.9062 | 0.7981 |
+| 0.0075        | 6.0   | 9882  | 0.8349          | 0.9109   | 0.9352 | 0.8866 | 0.9175 | 0.9407 | 0.9297 | 0.8219 |
+| 0.0           | 7.0   | 11529 | 0.8219          | 0.9160   | 0.9393 | 0.8927 | 0.9225 | 0.9412 | 0.9375 | 0.8321 |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c822e1ef018299535f5952678c64f19a070e3736ea6a741f912a7612dc91a69
+oid sha256:53ab6209cf4d8e7eef4b221cb99033bf5d01c478142d8523b042bdb2e58993d6
 size 433270768

runs/Apr14_21-08-14_tardis/events.out.tfevents.1713121729.tardis.128273.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4087eecff7227890b7d1877ec78856b2053003675ad196ce4bec2670baa2c19d
-size 11354
+oid sha256:2481241ffed35cb8f60cf104967905a64f6f8bfc7a006870c480e637b7a73e01
+size 14407

runs/Apr14_21-08-14_tardis/events.out.tfevents.1713123193.tardis.128273.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:168ead1ced6c580833f968d8977f96e81e3ebe3fa4ce15052beabfe2db5a8fee
+size 699