End of training

README.md

@@ -20,11 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [Clinical-AI-Apollo/Medical-NER](https://huggingface.co/Clinical-AI-Apollo/Medical-NER) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1886
-- Precision: 0.9483
-- Recall: 0.9420
-- F1: 0.9451
-- Accuracy: 0.9405
+- Loss: 0.2045
+- Precision: 0.9394
+- Recall: 0.9282
+- F1: 0.9338
+- Accuracy: 0.9296
 
 ## Model description
 
@@ -43,19 +43,37 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 2e-05
+- learning_rate: 2e-06
 - train_batch_size: 4
 - eval_batch_size: 4
 - seed: 42
-- gradient_accumulation_steps: 4
-- total_train_batch_size: 16
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 5
-- mixed_precision_training: Native AMP
+- num_epochs: 10
 
 ### Training results
 
+| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
+| No log        | 0.37  | 100  | 0.4486          | 0.8318    | 0.8662 | 0.8486 | 0.8331   |
+| No log        | 0.75  | 200  | 0.3747          | 0.8608    | 0.8834 | 0.8720 | 0.8646   |
+| No log        | 1.12  | 300  | 0.3245          | 0.8801    | 0.8932 | 0.8866 | 0.8828   |
+| No log        | 1.49  | 400  | 0.2846          | 0.9128    | 0.9038 | 0.9083 | 0.9028   |
+| 0.4808        | 1.87  | 500  | 0.2554          | 0.9199    | 0.9067 | 0.9133 | 0.9083   |
+| 0.4808        | 2.24  | 600  | 0.2480          | 0.9270    | 0.9073 | 0.9171 | 0.9102   |
+| 0.4808        | 2.61  | 700  | 0.2269          | 0.9271    | 0.9172 | 0.9221 | 0.9171   |
+| 0.4808        | 2.99  | 800  | 0.2319          | 0.9270    | 0.9089 | 0.9179 | 0.9129   |
+| 0.4808        | 3.36  | 900  | 0.2303          | 0.9284    | 0.9088 | 0.9185 | 0.9133   |
+| 0.2633        | 3.73  | 1000 | 0.2246          | 0.9311    | 0.9111 | 0.9210 | 0.9155   |
+| 0.2633        | 4.1   | 1100 | 0.2120          | 0.9343    | 0.9218 | 0.9280 | 0.9236   |
+| 0.2633        | 4.48  | 1200 | 0.2111          | 0.9361    | 0.9222 | 0.9291 | 0.9243   |
+| 0.2633        | 4.85  | 1300 | 0.2152          | 0.9320    | 0.9185 | 0.9252 | 0.9208   |
+| 0.2633        | 5.22  | 1400 | 0.2068          | 0.9333    | 0.9227 | 0.9280 | 0.9239   |
+| 0.2218        | 5.6   | 1500 | 0.2070          | 0.9360    | 0.9256 | 0.9308 | 0.9267   |
+| 0.2218        | 5.97  | 1600 | 0.2045          | 0.9394    | 0.9282 | 0.9338 | 0.9296   |
+| 0.2218        | 6.34  | 1700 | 0.2020          | 0.9357    | 0.9275 | 0.9316 | 0.9284   |
+| 0.2218        | 6.72  | 1800 | 0.2054          | 0.9354    | 0.9227 | 0.9290 | 0.9246   |
+| 0.2218        | 7.09  | 1900 | 0.2053          | 0.9372    | 0.9253 | 0.9312 | 0.9269   |
 
 
 ### Framework versions

config.json

@@ -11,8 +11,7 @@
     "0": "LABEL_0",
     "1": "LABEL_1",
     "2": "LABEL_2",
-    "3": "LABEL_3",
-    "4": "LABEL_4"
+    "3": "LABEL_3"
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
@@ -20,8 +19,7 @@
     "LABEL_0": 0,
     "LABEL_1": 1,
     "LABEL_2": 2,
-    "LABEL_3": 3,
-    "LABEL_4": 4
+    "LABEL_3": 3
   },
   "layer_norm_eps": 1e-07,
   "max_position_embeddings": 512,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6d5274078a776dd953b0390ee77fc2d3335ecd6e77093c3fa28017bef2033ef5
-size 735365964
+oid sha256:490e13939d69307a2a6946adb3aaa26d86c618f5272c46448ca8bc265bb0a0b5
+size 735362800

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:429f9a5397d21c04b6c82646ed585300c0105bd1539e5e6ae89c9bb49a39b6c5
+oid sha256:4ab3f3b39635cc5df4584fe1ac9aebb465b2a544014dd5cab2afd996c665ae77
 size 4920