procit011/anger_frustration_detection_9

README.md

@@ -19,9 +19,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [pdelobelle/robbert-v2-dutch-base](https://huggingface.co/pdelobelle/robbert-v2-dutch-base) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0817
-- F1: 0.9881
-- Accuracy: 0.9912
+- Loss: 0.0825
+- F1: 0.9840
+- Accuracy: 0.9881
+- F1 Yes: 0.976
 
 ## Model description
 
@@ -41,33 +42,30 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 1e-05
-- train_batch_size: 4
-- eval_batch_size: 8
+- train_batch_size: 8
+- eval_batch_size: 16
 - seed: 42
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 16
+- total_train_batch_size: 32
 - 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: cosine
 - lr_scheduler_warmup_steps: 0.1
-- num_epochs: 10
+- num_epochs: 5
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | F1     | Accuracy |
-|:-------------:|:-----:|:----:|:---------------:|:------:|:--------:|
-| 0.6276        | 1.0   | 1133 | 0.1196          | 0.9628 | 0.9726   |
-| 0.0555        | 2.0   | 2266 | 0.1409          | 0.9666 | 0.9757   |
-| 0.1979        | 3.0   | 3399 | 0.0880          | 0.9828 | 0.9872   |
-| 0.0280        | 4.0   | 4532 | 0.0820          | 0.9827 | 0.9872   |
-| 0.0041        | 5.0   | 5665 | 0.0911          | 0.9852 | 0.9890   |
-| 0.0005        | 6.0   | 6798 | 0.0922          | 0.9863 | 0.9898   |
-| 0.0000        | 7.0   | 7931 | 0.0859          | 0.9857 | 0.9894   |
-| 0.0000        | 8.0   | 9064 | 0.1041          | 0.9845 | 0.9885   |
+| Training Loss | Epoch | Step | Validation Loss | F1     | Accuracy | F1 Yes |
+|:-------------:|:-----:|:----:|:---------------:|:------:|:--------:|:------:|
+| 0.2779        | 1.0   | 567  | 0.1380          | 0.9687 | 0.9770   | 0.9525 |
+| 0.3814        | 2.0   | 1134 | 0.1375          | 0.9740 | 0.9810   | 0.9606 |
+| 0.2305        | 3.0   | 1701 | 0.1152          | 0.9820 | 0.9868   | 0.9728 |
+| 0.1595        | 4.0   | 2268 | 0.1057          | 0.9839 | 0.9881   | 0.9756 |
+| 0.1194        | 5.0   | 2835 | 0.1177          | 0.9832 | 0.9876   | 0.9746 |
 
 
 ### Framework versions
 
-- Transformers 5.0.0
-- Pytorch 2.10.0+cu128
+- Transformers 5.2.0
+- Pytorch 2.9.0+cu126
 - Datasets 4.0.0
 - Tokenizers 0.22.2

config.json

@@ -23,9 +23,8 @@
   "output_past": true,
   "pad_token_id": 1,
   "position_embedding_type": "absolute",
-  "problem_type": "single_label_classification",
   "tie_word_embeddings": true,
-  "transformers_version": "5.0.0",
+  "transformers_version": "5.2.0",
   "type_vocab_size": 1,
   "use_cache": false,
   "vocab_size": 40000

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6d55b0f90fcaa5b3ac5e813c7d602759d2fbc52dac07123909bc0084fcb48ca9
+oid sha256:371945cdbbc576156872474d9fa1f5070ead5abc9b079074ac9b3f4ce83e9b67
 size 467078720

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4aa3d4409b2aded889e0062791995a63e078af36de3110102a9db50a107514a5
+oid sha256:1ff208458410f00d03d161ad05ee53e9b03595a84035c1356eb962222cfd7af2
 size 5201