egujr001/swim2-base-model

README.md

@@ -0,0 +1,125 @@
+---
+license: apache-2.0
+base_model: microsoft/swinv2-tiny-patch4-window8-256
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: swim2-base-model
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# swim2-base-model
+
+This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3146
+- Accuracy: 0.9206
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 64
+- eval_batch_size: 64
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 3
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 0.5804        | 0.05  | 100  | 0.5395          | 0.7388   |
+| 0.2244        | 0.09  | 200  | 0.3057          | 0.8787   |
+| 0.134         | 0.14  | 300  | 0.2218          | 0.9129   |
+| 0.1786        | 0.18  | 400  | 0.1567          | 0.9373   |
+| 0.0924        | 0.23  | 500  | 0.1360          | 0.9464   |
+| 0.1217        | 0.27  | 600  | 0.1732          | 0.9349   |
+| 0.144         | 0.32  | 700  | 0.1233          | 0.9538   |
+| 0.0917        | 0.37  | 800  | 0.1655          | 0.9379   |
+| 0.1005        | 0.41  | 900  | 0.1047          | 0.9632   |
+| 0.1391        | 0.46  | 1000 | 0.1281          | 0.9554   |
+| 0.0488        | 0.5   | 1100 | 0.0965          | 0.9688   |
+| 0.0587        | 0.55  | 1200 | 0.1926          | 0.9460   |
+| 0.0827        | 0.59  | 1300 | 0.0982          | 0.9630   |
+| 0.035         | 0.64  | 1400 | 0.1011          | 0.9676   |
+| 0.0529        | 0.69  | 1500 | 0.0984          | 0.9646   |
+| 0.0653        | 0.73  | 1600 | 0.0877          | 0.9666   |
+| 0.0749        | 0.78  | 1700 | 0.1208          | 0.9604   |
+| 0.0686        | 0.82  | 1800 | 0.0742          | 0.9719   |
+| 0.039         | 0.87  | 1900 | 0.0829          | 0.9717   |
+| 0.0607        | 0.91  | 2000 | 0.0767          | 0.9746   |
+| 0.0478        | 0.96  | 2100 | 0.0789          | 0.9725   |
+| 0.0408        | 1.01  | 2200 | 0.0750          | 0.9757   |
+| 0.0228        | 1.05  | 2300 | 0.0707          | 0.9773   |
+| 0.0431        | 1.1   | 2400 | 0.0690          | 0.9787   |
+| 0.0675        | 1.14  | 2500 | 0.0712          | 0.9773   |
+| 0.0624        | 1.19  | 2600 | 0.1109          | 0.9640   |
+| 0.0843        | 1.23  | 2700 | 0.1077          | 0.9692   |
+| 0.0328        | 1.28  | 2800 | 0.0663          | 0.9795   |
+| 0.0724        | 1.33  | 2900 | 0.0811          | 0.9766   |
+| 0.0385        | 1.37  | 3000 | 0.0820          | 0.9732   |
+| 0.0315        | 1.42  | 3100 | 0.0711          | 0.9788   |
+| 0.0367        | 1.46  | 3200 | 0.0806          | 0.9765   |
+| 0.0382        | 1.51  | 3300 | 0.1444          | 0.9612   |
+| 0.024         | 1.55  | 3400 | 0.1038          | 0.9738   |
+| 0.0331        | 1.6   | 3500 | 0.1181          | 0.9660   |
+| 0.0419        | 1.65  | 3600 | 0.0687          | 0.9790   |
+| 0.0352        | 1.69  | 3700 | 0.0687          | 0.9789   |
+| 0.0588        | 1.74  | 3800 | 0.0620          | 0.9804   |
+| 0.0313        | 1.78  | 3900 | 0.0975          | 0.9722   |
+| 0.0421        | 1.83  | 4000 | 0.0588          | 0.9803   |
+| 0.0182        | 1.87  | 4100 | 0.0601          | 0.9819   |
+| 0.0323        | 1.92  | 4200 | 0.0593          | 0.9819   |
+| 0.0565        | 1.97  | 4300 | 0.0537          | 0.9820   |
+| 0.0266        | 2.01  | 4400 | 0.0693          | 0.9804   |
+| 0.0374        | 2.06  | 4500 | 0.0610          | 0.9819   |
+| 0.0246        | 2.1   | 4600 | 0.0580          | 0.9822   |
+| 0.0316        | 2.15  | 4700 | 0.0674          | 0.9804   |
+| 0.0415        | 2.19  | 4800 | 0.0569          | 0.9826   |
+| 0.0361        | 2.24  | 4900 | 0.0550          | 0.9840   |
+| 0.0298        | 2.29  | 5000 | 0.0575          | 0.9830   |
+| 0.0275        | 2.33  | 5100 | 0.0600          | 0.9836   |
+| 0.0194        | 2.38  | 5200 | 0.0678          | 0.9825   |
+| 0.0279        | 2.42  | 5300 | 0.0608          | 0.9838   |
+| 0.0585        | 2.47  | 5400 | 0.0548          | 0.9840   |
+| 0.0272        | 2.51  | 5500 | 0.0565          | 0.9841   |
+| 0.027         | 2.56  | 5600 | 0.0565          | 0.9840   |
+| 0.0154        | 2.61  | 5700 | 0.0671          | 0.9818   |
+| 0.0315        | 2.65  | 5800 | 0.0554          | 0.9851   |
+| 0.0351        | 2.7   | 5900 | 0.0638          | 0.9832   |
+| 0.0216        | 2.74  | 6000 | 0.0517          | 0.9851   |
+| 0.0218        | 2.79  | 6100 | 0.0574          | 0.9844   |
+| 0.0324        | 2.83  | 6200 | 0.0526          | 0.9851   |
+| 0.0365        | 2.88  | 6300 | 0.0546          | 0.9852   |
+| 0.0333        | 2.93  | 6400 | 0.0523          | 0.9854   |
+| 0.0121        | 2.97  | 6500 | 0.0570          | 0.9845   |
+
+
+### Framework versions
+
+- Transformers 4.33.0
+- Pytorch 2.0.0+cu117
+- Datasets 2.19.1
+- Tokenizers 0.13.3

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.9205868867491976,
+    "eval_loss": 0.3146313428878784,
+    "eval_runtime": 26.7942,
+    "eval_samples_per_second": 406.991,
+    "eval_steps_per_second": 6.382,
+    "total_flos": 1.366461317950577e+19,
+    "train_loss": 0.06796654012328891,
+    "train_runtime": 7965.932,
+    "train_samples_per_second": 52.725,
+    "train_steps_per_second": 0.824
+}

config.json

@@ -0,0 +1,54 @@
+{
+  "_name_or_path": "microsoft/swinv2-tiny-patch4-window8-256",
+  "architectures": [
+    "Swinv2ForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "depths": [
+    2,
+    2,
+    6,
+    2
+  ],
+  "drop_path_rate": 0.1,
+  "embed_dim": 96,
+  "encoder_stride": 32,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "Fake",
+    "1": "Real"
+  },
+  "image_size": 256,
+  "initializer_range": 0.02,
+  "label2id": {
+    "Fake": "0",
+    "Real": "1"
+  },
+  "layer_norm_eps": 1e-05,
+  "mlp_ratio": 4.0,
+  "model_type": "swinv2",
+  "num_channels": 3,
+  "num_heads": [
+    3,
+    6,
+    12,
+    24
+  ],
+  "num_layers": 4,
+  "patch_size": 4,
+  "path_norm": true,
+  "pretrained_window_sizes": [
+    0,
+    0,
+    0,
+    0
+  ],
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.33.0",
+  "use_absolute_embeddings": false,
+  "window_size": 8
+}

preprocessor_config.json

@@ -0,0 +1,23 @@
+{
+  "do_normalize": true,
+  "do_rescale": false,
+  "do_resize": true,
+  "feature_extractor_type": "ViTFeatureExtractor",
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_processor_type": "ViTImageProcessor",
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 256,
+    "width": 256
+  }
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a2e1908d2818ed1bee2bbb1be57ed0b5c3e9fc6d83199d880ccbdedaa94c7ff
+size 110405153

test_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.9205868867491976,
+    "eval_loss": 0.3146313428878784,
+    "eval_runtime": 26.7942,
+    "eval_samples_per_second": 406.991,
+    "eval_steps_per_second": 6.382
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.9933929515292639,
+    "eval_loss": 0.017951341345906258,
+    "eval_runtime": 352.1451,
+    "eval_samples_per_second": 397.569,
+    "eval_steps_per_second": 6.213
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af06d83af3497277e6f2bdb1ffd63d3d33286759bf27825794010a516729a63b
+size 4027

val_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.0,
+    "eval_accuracy": 0.9850997025448937,
+    "eval_loss": 0.05167504400014877,
+    "eval_runtime": 88.2766,
+    "eval_samples_per_second": 411.298,
+    "eval_steps_per_second": 6.434
+}