MultiPRIDE-LGBT-Baseline-it

README.md

@@ -21,11 +21,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [nickprock/setfit-italian-hate-speech](https://huggingface.co/nickprock/setfit-italian-hate-speech) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3546
-- Accuracy: 0.9693
-- F1: 0.9153
-- Precision: 0.9643
-- Recall: 0.8710
+- Loss: 0.6297
+- Accuracy: 0.9325
+- F1: 0.8831
+- Precision: 0.9131
+- Recall: 0.8596
 
 ## Model description
 
@@ -47,7 +47,7 @@ The following hyperparameters were used during training:
 - learning_rate: 2e-05
 - train_batch_size: 8
 - eval_batch_size: 8
-- seed: 67
+- seed: 150
 - optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - num_epochs: 10
@@ -56,14 +56,12 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | F1     | Precision | Recall |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
-| 0.5817        | 1.0   | 95   | 0.8524          | 0.8650   | 0.45   | 1.0       | 0.2903 |
-| 0.2861        | 2.0   | 190  | 0.2522          | 0.9571   | 0.8852 | 0.9       | 0.8710 |
-| 0.1572        | 3.0   | 285  | 0.2762          | 0.9632   | 0.8966 | 0.9630    | 0.8387 |
-| 0.0878        | 4.0   | 380  | 0.4025          | 0.9632   | 0.8966 | 0.9630    | 0.8387 |
-| 0.0754        | 5.0   | 475  | 0.3290          | 0.9693   | 0.9153 | 0.9643    | 0.8710 |
-| 0.0262        | 6.0   | 570  | 0.3922          | 0.9693   | 0.9153 | 0.9643    | 0.8710 |
-| 0.0003        | 7.0   | 665  | 0.4190          | 0.9693   | 0.9153 | 0.9643    | 0.8710 |
-| 0.0003        | 8.0   | 760  | 0.3546          | 0.9693   | 0.9153 | 0.9643    | 0.8710 |
+| 0.4607        | 1.0   | 95   | 0.3423          | 0.9264   | 0.8742 | 0.8965    | 0.8558 |
+| 0.2351        | 2.0   | 190  | 0.4198          | 0.9325   | 0.8831 | 0.9131    | 0.8596 |
+| 0.201         | 3.0   | 285  | 0.1992          | 0.9632   | 0.9430 | 0.9247    | 0.9649 |
+| 0.1004        | 4.0   | 380  | 0.4917          | 0.9202   | 0.8535 | 0.9155    | 0.8150 |
+| 0.1           | 5.0   | 475  | 0.3548          | 0.9387   | 0.8979 | 0.9086    | 0.8881 |
+| 0.0438        | 6.0   | 570  | 0.6297          | 0.9325   | 0.8831 | 0.9131    | 0.8596 |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6de507ac5b0c1d6cee6eca6de906d3db50cadadabed89bbf045aa0e136f963d7
+oid sha256:23c182b8e9fc6733c0ce59b25ea905ae1f4b13a3ba67a8a1a5ac811ab39d767c
 size 439740408

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9249ef562b0836084b17d4601c9da441aad645f6349b15813cb209b541ef5832
+oid sha256:63f596d70d03565682db6573d4aa92edb65e470eb2ce4231bedd7444253e4546
 size 5969