End of training

README.md

@@ -18,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/rembert](https://huggingface.co/google/rembert) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6926
+- Loss: 0.6811
 - Accuracy: 0.9205
 - 1-f1: 0.0
 - 1-recall: 0.0
@@ -48,16 +48,27 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.06
 - num_epochs: 20
 - mixed_precision_training: Native AMP
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Accuracy | 1-f1 | 1-recall | 1-precision | Balanced Acc |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|:----:|:--------:|:-----------:|:------------:|
-| 0.6947        | 1.0   | 82   | 0.6811          | 0.9205   | 0.0  | 0.0      | 0.0         | 0.5          |
-| 0.7535        | 2.0   | 164  | 0.6839          | 0.9205   | 0.0  | 0.0      | 0.0         | 0.5          |
-| 0.5308        | 3.0   | 246  | 0.6926          | 0.9205   | 0.0  | 0.0      | 0.0         | 0.5          |
+| Training Loss | Epoch | Step | Validation Loss | Accuracy | 1-f1   | 1-recall | 1-precision | Balanced Acc |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:--------:|:-----------:|:------------:|
+| 0.6923        | 1.0   | 82   | 0.6910          | 0.9144   | 0.0    | 0.0      | 0.0         | 0.4967       |
+| 0.5278        | 2.0   | 164  | 0.6997          | 0.9144   | 0.0    | 0.0      | 0.0         | 0.4967       |
+| 0.5443        | 3.0   | 246  | 0.9550          | 0.9205   | 0.0    | 0.0      | 0.0         | 0.5          |
+| 0.6379        | 4.0   | 328  | 0.6841          | 0.9052   | 0.0606 | 0.0385   | 0.1429      | 0.5093       |
+| 0.6825        | 5.0   | 410  | 0.6811          | 0.9205   | 0.0    | 0.0      | 0.0         | 0.5          |
+| 0.651         | 6.0   | 492  | 0.6834          | 0.9205   | 0.0    | 0.0      | 0.0         | 0.5          |
+| 0.619         | 7.0   | 574  | 0.6941          | 0.9205   | 0.0    | 0.0      | 0.0         | 0.5          |
+| 0.7377        | 8.0   | 656  | 0.6775          | 0.9205   | 0.0    | 0.0      | 0.0         | 0.5          |
+| 0.6771        | 9.0   | 738  | 0.6895          | 0.9205   | 0.0    | 0.0      | 0.0         | 0.5          |
+| 0.5212        | 10.0  | 820  | 0.7033          | 0.9205   | 0.0    | 0.0      | 0.0         | 0.5          |
+| 0.631         | 11.0  | 902  | 0.6980          | 0.9205   | 0.0    | 0.0      | 0.0         | 0.5          |
+| 0.5951        | 12.0  | 984  | 0.6824          | 0.9205   | 0.0    | 0.0      | 0.0         | 0.5          |
+| 1.0367        | 13.0  | 1066 | 0.6811          | 0.9205   | 0.0    | 0.0      | 0.0         | 0.5          |
 
 
 ### Framework versions

config.json

@@ -3,14 +3,14 @@
   "architectures": [
     "RemBertForSequenceClassification"
   ],
-  "attention_probs_dropout_prob": 0,
+  "attention_probs_dropout_prob": 0.2,
   "bos_token_id": 312,
   "classifier_dropout_prob": 0.1,
   "embedding_dropout_prob": 0,
   "embedding_size": 256,
   "eos_token_id": 313,
   "hidden_act": "gelu",
-  "hidden_dropout_prob": 0,
+  "hidden_dropout_prob": 0.2,
   "hidden_size": 1152,
   "initializer_range": 0.02,
   "input_embedding_size": 256,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a6de7dfd4b093ac567c520dc27222660cd2694f7e0b757ba44ab69c8616a2758
+oid sha256:07e375b11b570a948b59422a1235f3ce998e585bdba2e98edc924b62c36baf6a
 size 2303755328

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1fb2169ee5f1d5e073bf72fe8e9ec68d1deb4da7bf393f774bffdf6e196e0636
+oid sha256:a5e3774cf2ece399ac7088ab7454aca72f4e064cf1edf80ab3df4e42f58405b5
 size 5432