itsLeen/swin-large-ai-or-nott

README.md

@@ -23,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.9625
+      value: 0.98
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,8 +33,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [itsLeen/swin-large-ai-or-not](https://huggingface.co/itsLeen/swin-large-ai-or-not) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2540
-- Accuracy: 0.9625
+- Loss: 0.1841
+- Accuracy: 0.98
 
 ## Model description
 
@@ -54,13 +54,14 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 3e-05
-- train_batch_size: 8
+- train_batch_size: 16
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 16
-- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-06
 - lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 1000
 - num_epochs: 20
 - mixed_precision_training: Native AMP
 
@@ -68,14 +69,11 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.1485        | 1.0   | 100  | 0.2140          | 0.93     |
-| 0.0185        | 2.0   | 200  | 0.1999          | 0.9625   |
-| 0.0122        | 3.0   | 300  | 0.1703          | 0.9575   |
-| 0.0044        | 4.0   | 400  | 0.1952          | 0.9575   |
-| 0.0           | 5.0   | 500  | 0.2642          | 0.965    |
-| 0.0           | 6.0   | 600  | 0.2600          | 0.965    |
-| 0.0           | 7.0   | 700  | 0.3664          | 0.9625   |
-| 0.0           | 8.0   | 800  | 0.2540          | 0.9625   |
+| 0.0095        | 4.0   | 100  | 0.2067          | 0.97     |
+| 0.0013        | 8.0   | 200  | 0.1890          | 0.9725   |
+| 0.0014        | 12.0  | 300  | 0.2007          | 0.9725   |
+| 0.0003        | 16.0  | 400  | 0.1674          | 0.98     |
+| 0.0001        | 20.0  | 500  | 0.1841          | 0.98     |
 
 
 ### Framework versions

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 8.0,
-    "total_flos": 1.0028231120388096e+18,
-    "train_loss": 0.03964285453781485,
-    "train_runtime": 2016.2985,
-    "train_samples_per_second": 15.871,
-    "train_steps_per_second": 0.992
+    "epoch": 20.0,
+    "total_flos": 2.507057780097024e+18,
+    "train_loss": 0.003469271097332239,
+    "train_runtime": 3906.4221,
+    "train_samples_per_second": 8.192,
+    "train_steps_per_second": 0.128
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c49b17008b5fa8e6fe35e45d2bff15f60ac98f912bb0cd32bde3187b637ef6b2
+oid sha256:4370b272790dfd2ba592e3834f62e097c9b0bea64300d0b6372eea1e429d7125
 size 347498816

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 8.0,
-    "total_flos": 1.0028231120388096e+18,
-    "train_loss": 0.03964285453781485,
-    "train_runtime": 2016.2985,
-    "train_samples_per_second": 15.871,
-    "train_steps_per_second": 0.992
+    "epoch": 20.0,
+    "total_flos": 2.507057780097024e+18,
+    "train_loss": 0.003469271097332239,
+    "train_runtime": 3906.4221,
+    "train_samples_per_second": 8.192,
+    "train_steps_per_second": 0.128
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f71648cc0aa416296e2b8c942150816176f175ccd7e9d3c4e9dda8976c83f8d
-size 5176
+oid sha256:96fbff71158aeaea6c5cc48f551eea09142f4e372203f1dd2c98b1b7fde00973
+size 5112