Add model trained on 8 emotions

README.md

@@ -0,0 +1,91 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: distilbert-base-uncased
+tags:
+- generated_from_trainer
+model-index:
+- name: emotion-analysis-8-categories
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# emotion-analysis-8-categories
+
+This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.1962
+- F1 Macro: 0.6719
+- Precision Macro: 0.7346
+- Recall Macro: 0.6242
+- F1 Joy: 0.8597
+- Precision Joy: 0.8918
+- Recall Joy: 0.8299
+- F1 Trust: 0.7831
+- Precision Trust: 0.7855
+- Recall Trust: 0.7808
+- F1 Fear: 0.6737
+- Precision Fear: 0.6957
+- Recall Fear: 0.6531
+- F1 Surprise: 0.5475
+- Precision Surprise: 0.6968
+- Recall Surprise: 0.4509
+- F1 Sadness: 0.6627
+- Precision Sadness: 0.7585
+- Recall Sadness: 0.5884
+- F1 Disgust: 0.5613
+- Precision Disgust: 0.6485
+- Recall Disgust: 0.4948
+- F1 Anger: 0.6058
+- Precision Anger: 0.6950
+- Recall Anger: 0.5369
+- F1 Anticipation: 0.6813
+- Precision Anticipation: 0.7050
+- Recall Anticipation: 0.6593
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 16
+- eval_batch_size: 64
+- seed: 42
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 32
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- num_epochs: 3
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | F1 Macro | Precision Macro | Recall Macro | F1 Joy | Precision Joy | Recall Joy | F1 Trust | Precision Trust | Recall Trust | F1 Fear | Precision Fear | Recall Fear | F1 Surprise | Precision Surprise | Recall Surprise | F1 Sadness | Precision Sadness | Recall Sadness | F1 Disgust | Precision Disgust | Recall Disgust | F1 Anger | Precision Anger | Recall Anger | F1 Anticipation | Precision Anticipation | Recall Anticipation |
+|:-------------:|:------:|:----:|:---------------:|:--------:|:---------------:|:------------:|:------:|:-------------:|:----------:|:--------:|:---------------:|:------------:|:-------:|:--------------:|:-----------:|:-----------:|:------------------:|:---------------:|:----------:|:-----------------:|:--------------:|:----------:|:-----------------:|:--------------:|:--------:|:---------------:|:------------:|:---------------:|:----------------------:|:-------------------:|
+| 0.2317        | 0.5230 | 500  | 0.2208          | 0.5067   | 0.6799          | 0.4299       | 0.8305 | 0.9098        | 0.7640     | 0.7647   | 0.8             | 0.7323       | 0.0     | 0.0            | 0.0         | 0.3661      | 0.7851             | 0.2387          | 0.6069     | 0.7276            | 0.5205         | 0.2965     | 0.7474            | 0.1849         | 0.5658   | 0.6824          | 0.4833       | 0.6228          | 0.7872                 | 0.5153              |
+| 0.2054        | 1.0460 | 1000 | 0.1998          | 0.6436   | 0.7371          | 0.5768       | 0.8480 | 0.8991        | 0.8024     | 0.7686   | 0.8242          | 0.7200       | 0.5909  | 0.7324         | 0.4952      | 0.4663      | 0.6730             | 0.3568          | 0.6476     | 0.7299            | 0.5821         | 0.5434     | 0.5987            | 0.4974         | 0.6084   | 0.7059          | 0.5345       | 0.6756          | 0.7338                 | 0.6260              |
+| 0.1845        | 1.5690 | 1500 | 0.1958          | 0.6619   | 0.7507          | 0.5990       | 0.8444 | 0.8964        | 0.7980     | 0.7738   | 0.7977          | 0.7512       | 0.6851  | 0.8158         | 0.5905      | 0.4928      | 0.6861             | 0.3844          | 0.6592     | 0.7204            | 0.6077         | 0.5        | 0.65              | 0.4062         | 0.6509   | 0.6917          | 0.6147       | 0.6893          | 0.7478                 | 0.6393              |
+| 0.1772        | 2.0921 | 2000 | 0.1937          | 0.6690   | 0.7468          | 0.6086       | 0.8499 | 0.8925        | 0.8112     | 0.7784   | 0.8305          | 0.7323       | 0.6809  | 0.7711         | 0.6095      | 0.5285      | 0.6567             | 0.4422          | 0.6611     | 0.7284            | 0.6051         | 0.5303     | 0.6341            | 0.4557         | 0.6480   | 0.6797          | 0.6192       | 0.6746          | 0.7814                 | 0.5935              |
+| 0.1645        | 2.6151 | 2500 | 0.1941          | 0.6740   | 0.7413          | 0.6218       | 0.8472 | 0.9           | 0.8002     | 0.7839   | 0.8             | 0.7685       | 0.6772  | 0.7619         | 0.6095      | 0.5392      | 0.6729             | 0.4497          | 0.6695     | 0.7339            | 0.6154         | 0.5360     | 0.6506            | 0.4557         | 0.6499   | 0.7039          | 0.6036       | 0.6888          | 0.7068                 | 0.6718              |
+
+
+### Framework versions
+
+- Transformers 4.48.3
+- Pytorch 2.5.1+cu124
+- Datasets 3.3.2
+- Tokenizers 0.21.0

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b513cc009f72435b65529a6864048c3277abe711e4ed65966c2571bb2c977d06
+oid sha256:76a66ded82e03425c305eba6d406cea3fadc551f8e6d777e0f9473f6db96bb87
 size 267851024