Model save

README.md

@@ -16,16 +16,16 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert-base-multilingual-cased
 
-This model is a fine-tuned version of [google-bert/bert-base-multilingual-cased](https://huggingface.co/google-bert/bert-base-multilingual-cased) on an unknown dataset.
+This model is a fine-tuned version of [google-bert/bert-base-multilingual-cased](https://huggingface.co/google-bert/bert-base-multilingual-cased) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.3146
-- F1 Macro: 0.8469
-- F1: 0.8821
-- F1 Neg: 0.8117
-- Acc: 0.855
-- Prec: 0.9079
-- Recall: 0.8577
-- Mcc: 0.6961
+- Loss: 0.4995
+- F1 Macro: 0.8801
+- F1: 0.9146
+- F1 Neg: 0.8456
+- Acc: 0.89
+- Prec: 0.9272
+- Recall: 0.9023
+- Mcc: 0.7608
 
 ## Model description
 
@@ -51,28 +51,16 @@ The following hyperparameters were used during training:
 - distributed_type: multi-GPU
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 15
+- num_epochs: 3
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1     | F1 Neg | Acc    | Prec   | Recall | Mcc    |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:------:|:------:|:------:|:------:|
-| No log        | 1.0   | 400  | 0.5484          | 0.7319   | 0.8512 | 0.6126 | 0.785  | 0.7593 | 0.9685 | 0.5329 |
-| 0.4785        | 2.0   | 800  | 0.5584          | 0.8144   | 0.8799 | 0.7490 | 0.8375 | 0.8293 | 0.9370 | 0.6431 |
-| 0.3306        | 3.0   | 1200 | 0.6227          | 0.8174   | 0.8699 | 0.7649 | 0.8325 | 0.8582 | 0.8819 | 0.6354 |
-| 0.2531        | 4.0   | 1600 | 0.8030          | 0.8127   | 0.8773 | 0.7481 | 0.835  | 0.8310 | 0.9291 | 0.6370 |
-| 0.1264        | 5.0   | 2000 | 0.8665          | 0.8175   | 0.8740 | 0.7609 | 0.835  | 0.8481 | 0.9016 | 0.6381 |
-| 0.1264        | 6.0   | 2400 | 0.8824          | 0.8180   | 0.8776 | 0.7584 | 0.8375 | 0.8412 | 0.9173 | 0.6426 |
-| 0.0643        | 7.0   | 2800 | 1.1166          | 0.8279   | 0.8821 | 0.7737 | 0.845  | 0.8529 | 0.9134 | 0.6599 |
-| 0.0526        | 8.0   | 3200 | 1.2699          | 0.7936   | 0.8741 | 0.7131 | 0.825  | 0.8046 | 0.9567 | 0.6185 |
-| 0.0266        | 9.0   | 3600 | 1.2158          | 0.8247   | 0.8764 | 0.7730 | 0.84   | 0.8598 | 0.8937 | 0.6507 |
-| 0.0131        | 10.0  | 4000 | 1.2718          | 0.8288   | 0.8772 | 0.7805 | 0.8425 | 0.8687 | 0.8858 | 0.6580 |
-| 0.0131        | 11.0  | 4400 | 1.3405          | 0.8335   | 0.8859 | 0.7810 | 0.85   | 0.8566 | 0.9173 | 0.6710 |
-| 0.0087        | 12.0  | 4800 | 1.3709          | 0.8298   | 0.8847 | 0.7749 | 0.8475 | 0.8509 | 0.9213 | 0.6652 |
-| 0.0078        | 13.0  | 5200 | 1.3145          | 0.8387   | 0.8859 | 0.7915 | 0.8525 | 0.8707 | 0.9016 | 0.6784 |
-| 0.0036        | 14.0  | 5600 | 1.4501          | 0.8198   | 0.8801 | 0.7594 | 0.84   | 0.8393 | 0.9252 | 0.6482 |
-| 0.0049        | 15.0  | 6000 | 1.4557          | 0.8254   | 0.8839 | 0.7669 | 0.845  | 0.8429 | 0.9291 | 0.6595 |
+| 0.5806        | 1.0   | 1681 | 0.6537          | 0.8073   | 0.8716 | 0.7430 | 0.8287 | 0.8348 | 0.9118 | 0.6215 |
+| 0.4629        | 2.0   | 3362 | 0.4867          | 0.8458   | 0.8946 | 0.7971 | 0.8612 | 0.8674 | 0.9235 | 0.6952 |
+| 0.3256        | 3.0   | 5043 | 0.6003          | 0.8484   | 0.8946 | 0.8022 | 0.8625 | 0.8745 | 0.9157 | 0.6986 |
 
 
 ### Framework versions

runs/Mar25_23-14-58_tardis/events.out.tfevents.1711405399.tardis.885572.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b164eb81a64d08f8f0eeaf75b093d62f56ce7755e55ec7745478791d49b18ead
+size 699