alenatz/relation-extraction-bert-biocause

README.md

@@ -18,15 +18,16 @@ should probably proofread and complete it, then remove this comment. -->
 
 # relation-bert-biocause
 
-This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on the None dataset.
+This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0342
-- Precision: 0.6948
-- Recall: 0.4193
-- F1: 0.4710
-- Accuracy: 0.9927
-- Balanced Accuracy: 0.4193
-- Kappa: 0.2100
+- Loss: 0.2130
+- Precision: 0.1019
+- Recall: 0.5855
+- F1: 0.1737
+- Accuracy: 0.9399
+- Relation P: 0.1019
+- Relation R: 0.5855
+- Relation F1: 0.1737
 
 ## Model description
 
@@ -45,26 +46,30 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 8
-- eval_batch_size: 8
+- learning_rate: 4e-05
+- train_batch_size: 16
+- eval_batch_size: 16
 - seed: 42
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 16
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 1
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss | Precision | Recall | F1     | Accuracy | Balanced Accuracy | Kappa  |
-|:-------------:|:------:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|:-----------------:|:------:|
-| 0.0776        | 0.5764 | 100  | 0.0367          | 0.7530    | 0.3905 | 0.4323 | 0.9927   | 0.3905            | 0.1475 |
+| Training Loss | Epoch  | Step | Validation Loss | Precision | Recall | F1     | Accuracy | Relation P | Relation R | Relation F1 |
+|:-------------:|:------:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|:----------:|:----------:|:-----------:|
+| 0.7103        | 0.1282 | 20   | 0.3074          | 0.0214    | 0.2368 | 0.0392 | 0.8048   | 0.0214     | 0.2368     | 0.0392      |
+| 0.7103        | 0.2564 | 40   | 0.2230          | 0.0523    | 0.3882 | 0.0922 | 0.8985   | 0.0523     | 0.3882     | 0.0922      |
+| 0.7103        | 0.3846 | 60   | 0.2568          | 0.0983    | 0.5987 | 0.1688 | 0.9413   | 0.0983     | 0.5987     | 0.1688      |
+| 0.7103        | 0.5128 | 80   | 0.2166          | 0.0593    | 0.4671 | 0.1053 | 0.9000   | 0.0593     | 0.4671     | 0.1053      |
+| 0.7103        | 0.6410 | 100  | 0.2308          | 0.1240    | 0.6842 | 0.2099 | 0.9489   | 0.1240     | 0.6842     | 0.2099      |
+| 0.7103        | 0.7692 | 120  | 0.2246          | 0.1080    | 0.625  | 0.1841 | 0.9435   | 0.1080     | 0.625      | 0.1841      |
+| 0.7103        | 0.8974 | 140  | 0.2290          | 0.1196    | 0.6316 | 0.2010 | 0.9483   | 0.1196     | 0.6316     | 0.2010      |
 
 
 ### Framework versions
 
-- Transformers 4.41.2
+- Transformers 4.42.4
 - Pytorch 2.3.0+cu121
 - Datasets 2.20.0
 - Tokenizers 0.19.1

config.json

@@ -1,7 +1,7 @@
 {
   "_name_or_path": "bert-base-cased",
   "architectures": [
-    "BertForTokenClassification"
+    "BertForUnbalancedTokenClassification"
   ],
   "attention_probs_dropout_prob": 0.1,
   "classifier_dropout": null,
@@ -11,14 +11,12 @@
   "hidden_size": 768,
   "id2label": {
     "0": "O",
-    "1": "B-RELATION",
-    "2": "I-RELATION"
+    "1": "I-REL"
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
-    "B-RELATION": 1,
-    "I-RELATION": 2,
+    "I-REL": 1,
     "O": 0
   },
   "layer_norm_eps": 1e-12,
@@ -29,7 +27,7 @@
   "pad_token_id": 0,
   "position_embedding_type": "absolute",
   "torch_dtype": "float32",
-  "transformers_version": "4.41.2",
+  "transformers_version": "4.42.4",
   "type_vocab_size": 2,
   "use_cache": true,
   "vocab_size": 28996

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e67f29a0634d271027b3f82bebaadf272e947bb474c5967785313e09405d6561
-size 430911284
+oid sha256:9fe954b1fbb2265c59319c44afb705da8fb3722ce34294dc2f02df1f8ac0c8ef
+size 433270744

runs/Jul13_20-07-13_a518198a24e3/events.out.tfevents.1720901242.a518198a24e3.2562.12

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af99f53483db3667eade388ebef84ed854dad1c89dfbc573629131440804b8a5
+size 9876

runs/Jul13_20-07-13_a518198a24e3/events.out.tfevents.1720901362.a518198a24e3.2562.13

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d692ce1963f1ad64446c67f95e536f1724541d7c331a4b3764f61ebe34c9646d
+size 723

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba7bd6094f1df4848a18ad7ba9f68fdc4b8a36ebf0555f908045cd92efd744c1
-size 5048
+oid sha256:bd0a710f1e0519b4d45d49708236592299e5ce368a7b6ba8a193961e147b7a07
+size 5112