End of training

README.md

@@ -8,6 +8,9 @@ datasets:
 - audiofolder
 metrics:
 - accuracy
+- f1
+- precision
+- recall
 model-index:
 - name: distilhubert-finetuned-cry-detector
   results:
@@ -23,7 +26,16 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.9559471365638766
+      value: 0.9691629955947136
+    - name: F1
+      type: f1
+      value: 0.9692159230090303
+    - name: Precision
+      type: precision
+      value: 0.969310997758714
+    - name: Recall
+      type: recall
+      value: 0.9691629955947136
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,8 +45,11 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [ntu-spml/distilhubert](https://huggingface.co/ntu-spml/distilhubert) on the audiofolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1355
-- Accuracy: 0.9559
+- Loss: 0.0944
+- Accuracy: 0.9692
+- F1: 0.9692
+- Precision: 0.9693
+- Recall: 0.9692
 
 ## Model description
 
@@ -56,22 +71,23 @@ The following hyperparameters were used during training:
 - learning_rate: 0.0001
 - train_batch_size: 8
 - eval_batch_size: 8
-- seed: 123
+- seed: 42
 - gradient_accumulation_steps: 8
 - total_train_batch_size: 64
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.01
 - num_epochs: 5
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss | Accuracy |
-|:-------------:|:------:|:----:|:---------------:|:--------:|
-| No log        | 0.9825 | 14   | 0.1557          | 0.9559   |
-| No log        | 1.9649 | 28   | 0.1813          | 0.9339   |
-| No log        | 2.9474 | 42   | 0.1382          | 0.9515   |
-| No log        | 4.0    | 57   | 0.1407          | 0.9471   |
-| No log        | 4.9123 | 70   | 0.1355          | 0.9559   |
+| Training Loss | Epoch  | Step | Validation Loss | Accuracy | F1     | Precision | Recall |
+|:-------------:|:------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|
+| No log        | 0.9825 | 14   | 0.1775          | 0.9427   | 0.9434 | 0.9459    | 0.9427 |
+| No log        | 1.9649 | 28   | 0.1464          | 0.9515   | 0.9519 | 0.9533    | 0.9515 |
+| No log        | 2.9474 | 42   | 0.1139          | 0.9559   | 0.9556 | 0.9560    | 0.9559 |
+| No log        | 4.0    | 57   | 0.1042          | 0.9648   | 0.9649 | 0.9652    | 0.9648 |
+| No log        | 4.9123 | 70   | 0.0944          | 0.9692   | 0.9692 | 0.9693    | 0.9692 |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8bd8d7e669392ff9c2ece77ce63d0463a872df423d16b588f52882ca8ba65d38
+oid sha256:8ec8c96078435e2d6387b9c637a56592918ae756da25cd14314c7a53bf1ddc60
 size 94763496

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3bbc2ab399046469e48f55810dbf47021772d0111da2a9e27f3b2d23b8652f30
+oid sha256:5463051aa2696944e9d7acb1e707fc85f73738131d3b3e03a71ab783a7ee5bf9
 size 5176