jointriple/brand_classification_2_20240705_model_3

README.md

@@ -15,17 +15,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # results
 
-This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on an unknown dataset.
+This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 4.5664
-- Accuracy: 0.8187
-- F1 Macro: 0.4565
-- Accuracy Balanced: 0.4775
-- F1 Micro: 0.8187
-- Precision Macro: 0.4745
-- Recall Macro: 0.4775
-- Precision Micro: 0.8187
-- Recall Micro: 0.8187
+- Loss: 1.5162
+- Accuracy: 0.9213
 
 ## Model description
 
@@ -45,29 +38,24 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 2e-05
-- train_batch_size: 512
-- eval_batch_size: 512
+- train_batch_size: 128
+- eval_batch_size: 128
 - seed: 42
-- gradient_accumulation_steps: 4
-- total_train_batch_size: 2048
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 500
-- num_epochs: 8
-- mixed_precision_training: Native AMP
+- num_epochs: 7
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | Accuracy Balanced | F1 Micro | Precision Macro | Recall Macro | Precision Micro | Recall Micro |
-|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:-----------------:|:--------:|:---------------:|:------------:|:---------------:|:------------:|
-| 7.6422        | 1.0   | 148  | 7.6258          | 0.0007   | 0.0001   | 0.0010            | 0.0007   | 0.0002          | 0.0010       | 0.0007          | 0.0007       |
-| 7.4777        | 2.0   | 296  | 7.3871          | 0.0123   | 0.0006   | 0.0041            | 0.0123   | 0.0009          | 0.0041       | 0.0123          | 0.0123       |
-| 7.0699        | 2.99  | 444  | 6.9315          | 0.0698   | 0.0107   | 0.0208            | 0.0698   | 0.0153          | 0.0208       | 0.0698          | 0.0698       |
-| 6.5586        | 4.0   | 593  | 6.4103          | 0.2376   | 0.0462   | 0.0638            | 0.2376   | 0.0666          | 0.0638       | 0.2376          | 0.2376       |
-| 6.1134        | 5.0   | 741  | 5.9266          | 0.4651   | 0.1275   | 0.1506            | 0.4651   | 0.1543          | 0.1506       | 0.4651          | 0.4651       |
-| 5.6821        | 6.0   | 889  | 5.4735          | 0.6254   | 0.2306   | 0.2528            | 0.6254   | 0.2616          | 0.2528       | 0.6254          | 0.6254       |
-| 5.2577        | 6.99  | 1037 | 5.0254          | 0.7350   | 0.3385   | 0.3595            | 0.7350   | 0.3635          | 0.3595       | 0.7350          | 0.7350       |
-| 4.8083        | 7.99  | 1184 | 4.5664          | 0.8187   | 0.4565   | 0.4775            | 0.8187   | 0.4745          | 0.4775       | 0.8187          | 0.8187       |
+| Training Loss | Epoch | Step  | Validation Loss | Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:--------:|
+| 6.0895        | 1.0   | 2371  | 5.7708          | 0.4478   |
+| 4.5154        | 2.0   | 4742  | 4.3046          | 0.7988   |
+| 3.5146        | 3.0   | 7113  | 3.1486          | 0.8932   |
+| 2.6148        | 4.0   | 9484  | 2.3505          | 0.9134   |
+| 2.1514        | 5.0   | 11855 | 1.8621          | 0.9178   |
+| 1.9055        | 6.0   | 14226 | 1.5997          | 0.9210   |
+| 1.635         | 7.0   | 16597 | 1.5162          | 0.9213   |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:379f1ee1a9284c87a1ead7008c6fe3d75a805be95b81609c879cbe2980a05339
+oid sha256:a0747cc2f07db769db83f429e39b63f99aedfe36b2b73df178f2650855bd389a
 size 444369032

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:23e4b522a84b851d579dc66177ec184d2091c6a2b0722513709d9e48d683abc1
+oid sha256:c3ed22963347948e995ce143b94f0657e738f08389cbf6d8384707e6b51c37a3
 size 4856