Push ../models/xlnet/xlnet-base-cased/biored-augmentations-only/ trained on biored-train_160_splits.pt (160 samples)

README.md

@@ -28,12 +28,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [xlnet-base-cased](https://huggingface.co/xlnet-base-cased) on the bigbio/biored dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1515
-- Accuracy: 0.9536
-- Precision: 0.8724
-- Recall: 0.828
-- F1: 0.8467
-- Weighted F1: 0.9534
+- Loss: 0.1510
+- Accuracy: 0.9508
+- Precision: 0.8521
+- Recall: 0.8278
+- F1: 0.8391
+- Weighted F1: 0.9506
 
 ## Model description
 
@@ -52,26 +52,22 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 1.5e-05
+- learning_rate: 2e-05
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: cosine
+- lr_scheduler_type: linear
 - num_epochs: 50
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1     | Weighted F1 |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:---------:|:------:|:------:|:-----------:|
-| No log        | 1.0   | 20   | 0.2106          | 0.932    | 0.8372    | 0.7555 | 0.7921 | 0.9299      |
-| No log        | 2.0   | 40   | 0.2011          | 0.9358   | 0.798     | 0.8008 | 0.7972 | 0.9351      |
-| No log        | 3.0   | 60   | 0.1973          | 0.9411   | 0.8461    | 0.8033 | 0.8226 | 0.9403      |
-| No log        | 4.0   | 80   | 0.1949          | 0.9435   | 0.8337    | 0.8194 | 0.8259 | 0.9427      |
-| No log        | 5.0   | 100  | 0.2025          | 0.9437   | 0.8465    | 0.8156 | 0.8289 | 0.9433      |
-| No log        | 6.0   | 120  | 0.2026          | 0.9434   | 0.8348    | 0.8145 | 0.8235 | 0.9432      |
-| No log        | 7.0   | 140  | 0.2056          | 0.9459   | 0.8288    | 0.8082 | 0.8176 | 0.9451      |
-| No log        | 8.0   | 160  | 0.2087          | 0.9452   | 0.8137    | 0.8137 | 0.8126 | 0.9448      |
+| No log        | 1.0   | 20   | 0.2068          | 0.9335   | 0.8641    | 0.7475 | 0.7976 | 0.9312      |
+| No log        | 2.0   | 40   | 0.1962          | 0.939    | 0.8035    | 0.8046 | 0.8013 | 0.9382      |
+| No log        | 3.0   | 60   | 0.1965          | 0.9429   | 0.8654    | 0.7947 | 0.826  | 0.9415      |
+| No log        | 4.0   | 80   | 0.1964          | 0.9443   | 0.8279    | 0.8174 | 0.8218 | 0.9436      |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5a4cd192d0193e4299ccff40b98e634112ece2ddf126f9d374e5dc4e9aa482a5
+oid sha256:92740ba814283aa0031de770387afff16addd2c30207a78bbb762b5bab3ee2f9
 size 466917412

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8905bbf3407ef5149aa09f5be2f62367c8b09c0b632b7812282b7ff2c16cb00d
+oid sha256:283860a87c99ccc06e151dba087fa53ccab87e535f069b1dd6580b5cb6a32d52
 size 4219