BERTweet with correct label mapping

README.md

@@ -18,10 +18,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [vinai/bertweet-large](https://huggingface.co/vinai/bertweet-large) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5604
-- Accuracy: 0.7896
-- F1 Macro: 0.7852
-- F1 Weighted: 0.7897
+- Loss: 0.5607
+- Accuracy: 0.7885
+- F1 Macro: 0.7817
+- F1 Weighted: 0.7885
 
 ## Model description
 
@@ -46,7 +46,7 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 200
+- lr_scheduler_warmup_steps: 300
 - num_epochs: 2
 - mixed_precision_training: Native AMP
 
@@ -54,13 +54,13 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Weighted |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:-----------:|
-| 0.5941        | 1.0   | 1540 | 0.5823          | 0.7679   | 0.7565   | 0.7693      |
-| 0.3924        | 2.0   | 3080 | 0.5604          | 0.7896   | 0.7852   | 0.7897      |
+| 0.5943        | 1.0   | 1540 | 0.5735          | 0.7708   | 0.7592   | 0.7708      |
+| 0.3951        | 2.0   | 3080 | 0.5607          | 0.7885   | 0.7817   | 0.7885      |
 
 
 ### Framework versions
 
 - Transformers 5.0.0
-- Pytorch 2.9.0+cu128
+- Pytorch 2.10.0+cu128
 - Datasets 4.0.0
 - Tokenizers 0.22.2

config.json

@@ -13,21 +13,21 @@
   "hidden_dropout_prob": 0.1,
   "hidden_size": 1024,
   "id2label": {
-    "0": "Argumentative",
-    "1": "Expressive",
-    "2": "Informational",
-    "3": "Neutral",
-    "4": "Opinion"
+    "0": "Neutral",
+    "1": "Opinion",
+    "2": "Argumentative",
+    "3": "Expressive",
+    "4": "Informational"
   },
   "initializer_range": 0.02,
   "intermediate_size": 4096,
   "is_decoder": false,
   "label2id": {
-    "Argumentative": 0,
-    "Expressive": 1,
-    "Informational": 2,
-    "Neutral": 3,
-    "Opinion": 4
+    "Argumentative": 2,
+    "Expressive": 3,
+    "Informational": 4,
+    "Neutral": 0,
+    "Opinion": 1
   },
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 514,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e61b290db9c7a7e4a0cafc84a7cab969640e735d72941d45e96027545de258e4
+oid sha256:39ab1084b0f9a628d6656528e490667561772d8ca0a389e5923e6a745357e1bf
 size 1421507660

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:130f9fca7fe299b916db1ae1f0b91ec700c604536848511b6305c6081c69837d
+oid sha256:3f03248de6d4f0df254777fc607cccfd915f4a794a519828ee4735e4eaaa7958
 size 5265