mehdie/fine_tuned_emBERT

README.md

@@ -19,11 +19,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [morten-j/Mehdie_Extended-mBERT](https://huggingface.co/morten-j/Mehdie_Extended-mBERT) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2085
-- F1: 0.6316
-- F5: 0.6072
-- Precision: 0.7059
-- Recall: 0.5714
+- Loss: 0.2002
+- F1: 0.6452
+- F5: 0.6077
+- Precision: 0.7692
+- Recall: 0.5556
 
 ## Model description
 
@@ -49,23 +49,28 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.2
-- num_epochs: 10
+- num_epochs: 15
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | F1     | F5     | Precision | Recall |
 |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:---------:|:------:|
-| No log        | 1.0   | 38   | 0.2641          | 0.0    | 0.0    | 0.0       | 0.0    |
-| No log        | 2.0   | 76   | 0.2444          | 0.0    | 0.0    | 0.0       | 0.0    |
-| No log        | 3.0   | 114  | 0.2411          | 0.0    | 0.0    | 0.0       | 0.0    |
-| No log        | 4.0   | 152  | 0.2419          | 0.0    | 0.0    | 0.0       | 0.0    |
-| No log        | 5.0   | 190  | 0.1799          | 0.4375 | 0.3994 | 0.5833    | 0.35   |
-| No log        | 6.0   | 228  | 0.2493          | 0.4444 | 0.4641 | 0.4       | 0.5    |
-| No log        | 7.0   | 266  | 0.2708          | 0.4444 | 0.4641 | 0.4       | 0.5    |
-| No log        | 8.0   | 304  | 0.2356          | 0.5517 | 0.4819 | 0.8889    | 0.4    |
-| No log        | 9.0   | 342  | 0.2139          | 0.5333 | 0.4731 | 0.8       | 0.4    |
-| No log        | 10.0  | 380  | 0.2477          | 0.4444 | 0.3754 | 0.8571    | 0.3    |
+| No log        | 1.0   | 38   | 0.2621          | 0.0    | 0.0    | 0.0       | 0.0    |
+| No log        | 2.0   | 76   | 0.2476          | 0.0    | 0.0    | 0.0       | 0.0    |
+| No log        | 3.0   | 114  | 0.1800          | 0.4286 | 0.3683 | 0.75      | 0.3    |
+| No log        | 4.0   | 152  | 0.1379          | 0.6667 | 0.6962 | 0.6       | 0.75   |
+| No log        | 5.0   | 190  | 0.1873          | 0.5625 | 0.5135 | 0.75      | 0.45   |
+| No log        | 6.0   | 228  | 0.1763          | 0.6857 | 0.6502 | 0.8       | 0.6    |
+| No log        | 7.0   | 266  | 0.1394          | 0.7895 | 0.7739 | 0.8333    | 0.75   |
+| No log        | 8.0   | 304  | 0.1685          | 0.7027 | 0.6816 | 0.7647    | 0.65   |
+| No log        | 9.0   | 342  | 0.2521          | 0.6    | 0.5322 | 0.9       | 0.45   |
+| No log        | 10.0  | 380  | 0.1923          | 0.7179 | 0.7110 | 0.7368    | 0.7    |
+| No log        | 11.0  | 418  | 0.1680          | 0.7027 | 0.6816 | 0.7647    | 0.65   |
+| No log        | 12.0  | 456  | 0.2524          | 0.5625 | 0.5135 | 0.75      | 0.45   |
+| No log        | 13.0  | 494  | 0.1693          | 0.6667 | 0.6395 | 0.75      | 0.6    |
+| 0.1001        | 14.0  | 532  | 0.1711          | 0.6667 | 0.6395 | 0.75      | 0.6    |
+| 0.1001        | 15.0  | 570  | 0.2301          | 0.7222 | 0.6928 | 0.8125    | 0.65   |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1c292e4be3c83eea7764cf7cd851a7d2a51b94a284d204995ad625f3ec82f28a
+oid sha256:feb02bb10f3235aff5f45362d7c9468f3ec690a67400e7a141ae2184b228535a
 size 669455360

runs/May16_08-43-10_yoga/events.out.tfevents.1715841790.yoga.7965.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac2c0a3028c9df74ab65169d3534365234e6520ebebcc6998f4fdd2530992675
+size 12386

runs/May16_08-43-10_yoga/events.out.tfevents.1715841922.yoga.7965.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dba8dff8ff711df4ea79aef03a3a9b24b81a8cf99885167b674b2b93cef5dcb
+size 554

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cf78b44e18668153d6ba469cc16394f3c6d1d51e64a3383c0a9cfa235df72b66
+oid sha256:ce14aca67f1bbcf3a566db2b0e9ea290c3fd47e8abab92795edd95264bb0d418
 size 4920