Initial model upload
Browse files- checkpoint-35008/config.json +32 -0
- checkpoint-35008/model.safetensors +3 -0
- checkpoint-35008/optimizer.pt +3 -0
- checkpoint-35008/preprocessor_config.json +22 -0
- checkpoint-35008/rng_state.pth +3 -0
- checkpoint-35008/scheduler.pt +3 -0
- checkpoint-35008/trainer_state.json +595 -0
- checkpoint-35008/training_args.bin +3 -0
- config.json +32 -0
- model.safetensors +3 -0
- preprocessor_config.json +22 -0
- training_args.bin +3 -0
checkpoint-35008/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vit-base-patch16-224-in21k",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"ViTForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"encoder_stride": 16,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.0,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "Real",
|
| 13 |
+
"1": "Fake"
|
| 14 |
+
},
|
| 15 |
+
"image_size": 224,
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"Fake": 1,
|
| 20 |
+
"Real": 0
|
| 21 |
+
},
|
| 22 |
+
"layer_norm_eps": 1e-12,
|
| 23 |
+
"model_type": "vit",
|
| 24 |
+
"num_attention_heads": 12,
|
| 25 |
+
"num_channels": 3,
|
| 26 |
+
"num_hidden_layers": 12,
|
| 27 |
+
"patch_size": 16,
|
| 28 |
+
"problem_type": "single_label_classification",
|
| 29 |
+
"qkv_bias": true,
|
| 30 |
+
"torch_dtype": "float32",
|
| 31 |
+
"transformers_version": "4.46.3"
|
| 32 |
+
}
|
checkpoint-35008/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a3b12bec784c2a5d63adda101dd9a2c52aa4c4409fa27fb3e8030a31c174a17c
|
| 3 |
+
size 343223968
|
checkpoint-35008/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1926583928afb356bcfd7e7c04173ee58b1e09750d4c08c899e386969cd12312
|
| 3 |
+
size 686568890
|
checkpoint-35008/preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.5,
|
| 7 |
+
0.5,
|
| 8 |
+
0.5
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessor",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
checkpoint-35008/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:815a0424805de0b5a0613e02d7f7f0eed2aed1068d83e0e6c5de0b39b39d53a0
|
| 3 |
+
size 14244
|
checkpoint-35008/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:77bf7dda0c245c4898536820dfd19cf4f54fab88acfc128f98172d49b925f26e
|
| 3 |
+
size 1064
|
checkpoint-35008/trainer_state.json
ADDED
|
@@ -0,0 +1,595 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.1280793398618698,
|
| 3 |
+
"best_model_checkpoint": "deepfake_vs_real_image_detection/checkpoint-35008",
|
| 4 |
+
"epoch": 8.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 35008,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.11425959780621572,
|
| 13 |
+
"grad_norm": 0.9279178977012634,
|
| 14 |
+
"learning_rate": 9.87127410034899e-07,
|
| 15 |
+
"loss": 0.6754,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.22851919561243145,
|
| 20 |
+
"grad_norm": 1.445857286453247,
|
| 21 |
+
"learning_rate": 9.72824532295898e-07,
|
| 22 |
+
"loss": 0.5645,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.34277879341864714,
|
| 27 |
+
"grad_norm": 2.294506788253784,
|
| 28 |
+
"learning_rate": 9.585216545568968e-07,
|
| 29 |
+
"loss": 0.3891,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.4570383912248629,
|
| 34 |
+
"grad_norm": 2.4269814491271973,
|
| 35 |
+
"learning_rate": 9.442187768178958e-07,
|
| 36 |
+
"loss": 0.2769,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.5712979890310786,
|
| 41 |
+
"grad_norm": 8.800315856933594,
|
| 42 |
+
"learning_rate": 9.299158990788947e-07,
|
| 43 |
+
"loss": 0.225,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.6855575868372943,
|
| 48 |
+
"grad_norm": 2.500453233718872,
|
| 49 |
+
"learning_rate": 9.156130213398935e-07,
|
| 50 |
+
"loss": 0.1974,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.79981718464351,
|
| 55 |
+
"grad_norm": 2.538663864135742,
|
| 56 |
+
"learning_rate": 9.013101436008924e-07,
|
| 57 |
+
"loss": 0.1816,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.9140767824497258,
|
| 62 |
+
"grad_norm": 3.2626090049743652,
|
| 63 |
+
"learning_rate": 8.870072658618913e-07,
|
| 64 |
+
"loss": 0.1687,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.0,
|
| 69 |
+
"eval_accuracy": 0.9183575124277163,
|
| 70 |
+
"eval_loss": 0.26723140478134155,
|
| 71 |
+
"eval_runtime": 613.6678,
|
| 72 |
+
"eval_samples_per_second": 64.25,
|
| 73 |
+
"eval_steps_per_second": 8.032,
|
| 74 |
+
"step": 4376
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.0283363802559415,
|
| 78 |
+
"grad_norm": 7.749739170074463,
|
| 79 |
+
"learning_rate": 8.727043881228902e-07,
|
| 80 |
+
"loss": 0.161,
|
| 81 |
+
"step": 4500
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.1425959780621573,
|
| 85 |
+
"grad_norm": 3.7298998832702637,
|
| 86 |
+
"learning_rate": 8.584015103838892e-07,
|
| 87 |
+
"loss": 0.148,
|
| 88 |
+
"step": 5000
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.2568555758683728,
|
| 92 |
+
"grad_norm": 1.4705345630645752,
|
| 93 |
+
"learning_rate": 8.440986326448881e-07,
|
| 94 |
+
"loss": 0.1456,
|
| 95 |
+
"step": 5500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.3711151736745886,
|
| 99 |
+
"grad_norm": 3.6541574001312256,
|
| 100 |
+
"learning_rate": 8.297957549058871e-07,
|
| 101 |
+
"loss": 0.1427,
|
| 102 |
+
"step": 6000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.4853747714808043,
|
| 106 |
+
"grad_norm": 2.981785535812378,
|
| 107 |
+
"learning_rate": 8.15492877166886e-07,
|
| 108 |
+
"loss": 0.1376,
|
| 109 |
+
"step": 6500
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.59963436928702,
|
| 113 |
+
"grad_norm": 5.706754684448242,
|
| 114 |
+
"learning_rate": 8.011899994278849e-07,
|
| 115 |
+
"loss": 0.1257,
|
| 116 |
+
"step": 7000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.7138939670932358,
|
| 120 |
+
"grad_norm": 4.674542427062988,
|
| 121 |
+
"learning_rate": 7.868871216888838e-07,
|
| 122 |
+
"loss": 0.1247,
|
| 123 |
+
"step": 7500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.8281535648994516,
|
| 127 |
+
"grad_norm": 10.191073417663574,
|
| 128 |
+
"learning_rate": 7.725842439498827e-07,
|
| 129 |
+
"loss": 0.124,
|
| 130 |
+
"step": 8000
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.9424131627056673,
|
| 134 |
+
"grad_norm": 8.716536521911621,
|
| 135 |
+
"learning_rate": 7.582813662108815e-07,
|
| 136 |
+
"loss": 0.1196,
|
| 137 |
+
"step": 8500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.0,
|
| 141 |
+
"eval_accuracy": 0.9366440093334686,
|
| 142 |
+
"eval_loss": 0.19802308082580566,
|
| 143 |
+
"eval_runtime": 414.2816,
|
| 144 |
+
"eval_samples_per_second": 95.172,
|
| 145 |
+
"eval_steps_per_second": 11.898,
|
| 146 |
+
"step": 8752
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.056672760511883,
|
| 150 |
+
"grad_norm": 0.8615439534187317,
|
| 151 |
+
"learning_rate": 7.439784884718805e-07,
|
| 152 |
+
"loss": 0.1161,
|
| 153 |
+
"step": 9000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.170932358318099,
|
| 157 |
+
"grad_norm": 4.869253158569336,
|
| 158 |
+
"learning_rate": 7.296756107328794e-07,
|
| 159 |
+
"loss": 0.1169,
|
| 160 |
+
"step": 9500
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.2851919561243146,
|
| 164 |
+
"grad_norm": 4.5460333824157715,
|
| 165 |
+
"learning_rate": 7.153727329938783e-07,
|
| 166 |
+
"loss": 0.1095,
|
| 167 |
+
"step": 10000
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 2.3994515539305303,
|
| 171 |
+
"grad_norm": 0.5965484976768494,
|
| 172 |
+
"learning_rate": 7.010698552548773e-07,
|
| 173 |
+
"loss": 0.1131,
|
| 174 |
+
"step": 10500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.5137111517367456,
|
| 178 |
+
"grad_norm": 2.817739725112915,
|
| 179 |
+
"learning_rate": 6.867669775158762e-07,
|
| 180 |
+
"loss": 0.1056,
|
| 181 |
+
"step": 11000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.6279707495429614,
|
| 185 |
+
"grad_norm": 2.231886625289917,
|
| 186 |
+
"learning_rate": 6.724640997768751e-07,
|
| 187 |
+
"loss": 0.1086,
|
| 188 |
+
"step": 11500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.742230347349177,
|
| 192 |
+
"grad_norm": 7.3367600440979,
|
| 193 |
+
"learning_rate": 6.58161222037874e-07,
|
| 194 |
+
"loss": 0.0982,
|
| 195 |
+
"step": 12000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.856489945155393,
|
| 199 |
+
"grad_norm": 1.7964383363723755,
|
| 200 |
+
"learning_rate": 6.438583442988729e-07,
|
| 201 |
+
"loss": 0.1018,
|
| 202 |
+
"step": 12500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.9707495429616086,
|
| 206 |
+
"grad_norm": 6.008702754974365,
|
| 207 |
+
"learning_rate": 6.295554665598718e-07,
|
| 208 |
+
"loss": 0.1044,
|
| 209 |
+
"step": 13000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 3.0,
|
| 213 |
+
"eval_accuracy": 0.9433904839200569,
|
| 214 |
+
"eval_loss": 0.17182812094688416,
|
| 215 |
+
"eval_runtime": 413.3537,
|
| 216 |
+
"eval_samples_per_second": 95.386,
|
| 217 |
+
"eval_steps_per_second": 11.924,
|
| 218 |
+
"step": 13128
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 3.0850091407678244,
|
| 222 |
+
"grad_norm": 8.11001205444336,
|
| 223 |
+
"learning_rate": 6.152525888208708e-07,
|
| 224 |
+
"loss": 0.0962,
|
| 225 |
+
"step": 13500
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 3.19926873857404,
|
| 229 |
+
"grad_norm": 3.732985019683838,
|
| 230 |
+
"learning_rate": 6.009497110818696e-07,
|
| 231 |
+
"loss": 0.0903,
|
| 232 |
+
"step": 14000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 3.313528336380256,
|
| 236 |
+
"grad_norm": 8.172978401184082,
|
| 237 |
+
"learning_rate": 5.866468333428685e-07,
|
| 238 |
+
"loss": 0.1025,
|
| 239 |
+
"step": 14500
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 3.4277879341864717,
|
| 243 |
+
"grad_norm": 7.130813121795654,
|
| 244 |
+
"learning_rate": 5.723439556038675e-07,
|
| 245 |
+
"loss": 0.0934,
|
| 246 |
+
"step": 15000
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.5420475319926874,
|
| 250 |
+
"grad_norm": 10.737488746643066,
|
| 251 |
+
"learning_rate": 5.580410778648664e-07,
|
| 252 |
+
"loss": 0.0937,
|
| 253 |
+
"step": 15500
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 3.656307129798903,
|
| 257 |
+
"grad_norm": 5.984869480133057,
|
| 258 |
+
"learning_rate": 5.437382001258653e-07,
|
| 259 |
+
"loss": 0.0947,
|
| 260 |
+
"step": 16000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 3.770566727605119,
|
| 264 |
+
"grad_norm": 1.0398708581924438,
|
| 265 |
+
"learning_rate": 5.294353223868642e-07,
|
| 266 |
+
"loss": 0.0944,
|
| 267 |
+
"step": 16500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 3.8848263254113347,
|
| 271 |
+
"grad_norm": 4.8392510414123535,
|
| 272 |
+
"learning_rate": 5.151324446478631e-07,
|
| 273 |
+
"loss": 0.0908,
|
| 274 |
+
"step": 17000
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 3.9990859232175504,
|
| 278 |
+
"grad_norm": 6.028433322906494,
|
| 279 |
+
"learning_rate": 5.00829566908862e-07,
|
| 280 |
+
"loss": 0.0897,
|
| 281 |
+
"step": 17500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 4.0,
|
| 285 |
+
"eval_accuracy": 0.9511514659632748,
|
| 286 |
+
"eval_loss": 0.1482020914554596,
|
| 287 |
+
"eval_runtime": 408.5773,
|
| 288 |
+
"eval_samples_per_second": 96.501,
|
| 289 |
+
"eval_steps_per_second": 12.064,
|
| 290 |
+
"step": 17504
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 4.113345521023766,
|
| 294 |
+
"grad_norm": 1.668658971786499,
|
| 295 |
+
"learning_rate": 4.86526689169861e-07,
|
| 296 |
+
"loss": 0.0867,
|
| 297 |
+
"step": 18000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.227605118829982,
|
| 301 |
+
"grad_norm": 3.3406217098236084,
|
| 302 |
+
"learning_rate": 4.7222381143085986e-07,
|
| 303 |
+
"loss": 0.0933,
|
| 304 |
+
"step": 18500
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 4.341864716636198,
|
| 308 |
+
"grad_norm": 6.337904930114746,
|
| 309 |
+
"learning_rate": 4.579209336918588e-07,
|
| 310 |
+
"loss": 0.0876,
|
| 311 |
+
"step": 19000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 4.456124314442413,
|
| 315 |
+
"grad_norm": 4.625348091125488,
|
| 316 |
+
"learning_rate": 4.436180559528577e-07,
|
| 317 |
+
"loss": 0.0857,
|
| 318 |
+
"step": 19500
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 4.570383912248629,
|
| 322 |
+
"grad_norm": 0.618135929107666,
|
| 323 |
+
"learning_rate": 4.293151782138566e-07,
|
| 324 |
+
"loss": 0.0875,
|
| 325 |
+
"step": 20000
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 4.684643510054845,
|
| 329 |
+
"grad_norm": 17.393951416015625,
|
| 330 |
+
"learning_rate": 4.1501230047485555e-07,
|
| 331 |
+
"loss": 0.0866,
|
| 332 |
+
"step": 20500
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 4.798903107861061,
|
| 336 |
+
"grad_norm": 0.8322166800498962,
|
| 337 |
+
"learning_rate": 4.0070942273585443e-07,
|
| 338 |
+
"loss": 0.088,
|
| 339 |
+
"step": 21000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 4.913162705667276,
|
| 343 |
+
"grad_norm": 0.31342649459838867,
|
| 344 |
+
"learning_rate": 3.864065449968533e-07,
|
| 345 |
+
"loss": 0.0884,
|
| 346 |
+
"step": 21500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 5.0,
|
| 350 |
+
"eval_accuracy": 0.9540681749010855,
|
| 351 |
+
"eval_loss": 0.13915510475635529,
|
| 352 |
+
"eval_runtime": 410.8966,
|
| 353 |
+
"eval_samples_per_second": 95.956,
|
| 354 |
+
"eval_steps_per_second": 11.996,
|
| 355 |
+
"step": 21880
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 5.027422303473492,
|
| 359 |
+
"grad_norm": 15.877787590026855,
|
| 360 |
+
"learning_rate": 3.7210366725785225e-07,
|
| 361 |
+
"loss": 0.0907,
|
| 362 |
+
"step": 22000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 5.141681901279708,
|
| 366 |
+
"grad_norm": 3.825509548187256,
|
| 367 |
+
"learning_rate": 3.578007895188512e-07,
|
| 368 |
+
"loss": 0.0784,
|
| 369 |
+
"step": 22500
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 5.255941499085923,
|
| 373 |
+
"grad_norm": 8.778756141662598,
|
| 374 |
+
"learning_rate": 3.434979117798501e-07,
|
| 375 |
+
"loss": 0.0839,
|
| 376 |
+
"step": 23000
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 5.3702010968921385,
|
| 380 |
+
"grad_norm": 3.6557865142822266,
|
| 381 |
+
"learning_rate": 3.29195034040849e-07,
|
| 382 |
+
"loss": 0.0834,
|
| 383 |
+
"step": 23500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 5.484460694698354,
|
| 387 |
+
"grad_norm": 4.403536319732666,
|
| 388 |
+
"learning_rate": 3.148921563018479e-07,
|
| 389 |
+
"loss": 0.0846,
|
| 390 |
+
"step": 24000
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 5.59872029250457,
|
| 394 |
+
"grad_norm": 1.8743895292282104,
|
| 395 |
+
"learning_rate": 3.005892785628468e-07,
|
| 396 |
+
"loss": 0.0818,
|
| 397 |
+
"step": 24500
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 5.712979890310786,
|
| 401 |
+
"grad_norm": 5.720378875732422,
|
| 402 |
+
"learning_rate": 2.8628640082384576e-07,
|
| 403 |
+
"loss": 0.0791,
|
| 404 |
+
"step": 25000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 5.8272394881170015,
|
| 408 |
+
"grad_norm": 9.525908470153809,
|
| 409 |
+
"learning_rate": 2.719835230848447e-07,
|
| 410 |
+
"loss": 0.0887,
|
| 411 |
+
"step": 25500
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 5.941499085923217,
|
| 415 |
+
"grad_norm": 13.019225120544434,
|
| 416 |
+
"learning_rate": 2.576806453458435e-07,
|
| 417 |
+
"loss": 0.0785,
|
| 418 |
+
"step": 26000
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 6.0,
|
| 422 |
+
"eval_accuracy": 0.9561479151871767,
|
| 423 |
+
"eval_loss": 0.13129685819149017,
|
| 424 |
+
"eval_runtime": 380.134,
|
| 425 |
+
"eval_samples_per_second": 103.721,
|
| 426 |
+
"eval_steps_per_second": 12.966,
|
| 427 |
+
"step": 26256
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 6.055758683729433,
|
| 431 |
+
"grad_norm": 8.515515327453613,
|
| 432 |
+
"learning_rate": 2.4337776760684246e-07,
|
| 433 |
+
"loss": 0.0781,
|
| 434 |
+
"step": 26500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 6.170018281535649,
|
| 438 |
+
"grad_norm": 2.822601318359375,
|
| 439 |
+
"learning_rate": 2.290748898678414e-07,
|
| 440 |
+
"loss": 0.0792,
|
| 441 |
+
"step": 27000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.2842778793418645,
|
| 445 |
+
"grad_norm": 0.4864025115966797,
|
| 446 |
+
"learning_rate": 2.147720121288403e-07,
|
| 447 |
+
"loss": 0.0789,
|
| 448 |
+
"step": 27500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 6.39853747714808,
|
| 452 |
+
"grad_norm": 1.266471266746521,
|
| 453 |
+
"learning_rate": 2.0046913438983924e-07,
|
| 454 |
+
"loss": 0.085,
|
| 455 |
+
"step": 28000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 6.512797074954296,
|
| 459 |
+
"grad_norm": 0.17314019799232483,
|
| 460 |
+
"learning_rate": 1.8616625665083812e-07,
|
| 461 |
+
"loss": 0.0771,
|
| 462 |
+
"step": 28500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 6.627056672760512,
|
| 466 |
+
"grad_norm": 0.11477702111005783,
|
| 467 |
+
"learning_rate": 1.7186337891183706e-07,
|
| 468 |
+
"loss": 0.083,
|
| 469 |
+
"step": 29000
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 6.7413162705667276,
|
| 473 |
+
"grad_norm": 3.228081703186035,
|
| 474 |
+
"learning_rate": 1.5756050117283597e-07,
|
| 475 |
+
"loss": 0.0815,
|
| 476 |
+
"step": 29500
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 6.855575868372943,
|
| 480 |
+
"grad_norm": 8.033062934875488,
|
| 481 |
+
"learning_rate": 1.4325762343383488e-07,
|
| 482 |
+
"loss": 0.0769,
|
| 483 |
+
"step": 30000
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 6.969835466179159,
|
| 487 |
+
"grad_norm": 3.3470427989959717,
|
| 488 |
+
"learning_rate": 1.289547456948338e-07,
|
| 489 |
+
"loss": 0.0827,
|
| 490 |
+
"step": 30500
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 7.0,
|
| 494 |
+
"eval_accuracy": 0.9570609718981434,
|
| 495 |
+
"eval_loss": 0.13065218925476074,
|
| 496 |
+
"eval_runtime": 375.1187,
|
| 497 |
+
"eval_samples_per_second": 105.108,
|
| 498 |
+
"eval_steps_per_second": 13.14,
|
| 499 |
+
"step": 30632
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 7.084095063985375,
|
| 503 |
+
"grad_norm": 3.7121076583862305,
|
| 504 |
+
"learning_rate": 1.1465186795583271e-07,
|
| 505 |
+
"loss": 0.0788,
|
| 506 |
+
"step": 31000
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 7.198354661791591,
|
| 510 |
+
"grad_norm": 1.4164518117904663,
|
| 511 |
+
"learning_rate": 1.0034899021683162e-07,
|
| 512 |
+
"loss": 0.0844,
|
| 513 |
+
"step": 31500
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 7.312614259597806,
|
| 517 |
+
"grad_norm": 0.26970207691192627,
|
| 518 |
+
"learning_rate": 8.604611247783053e-08,
|
| 519 |
+
"loss": 0.0795,
|
| 520 |
+
"step": 32000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 7.426873857404022,
|
| 524 |
+
"grad_norm": 2.8127174377441406,
|
| 525 |
+
"learning_rate": 7.174323473882945e-08,
|
| 526 |
+
"loss": 0.0846,
|
| 527 |
+
"step": 32500
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 7.541133455210238,
|
| 531 |
+
"grad_norm": 11.107344627380371,
|
| 532 |
+
"learning_rate": 5.744035699982836e-08,
|
| 533 |
+
"loss": 0.0797,
|
| 534 |
+
"step": 33000
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 7.655393053016454,
|
| 538 |
+
"grad_norm": 10.199967384338379,
|
| 539 |
+
"learning_rate": 4.3137479260827274e-08,
|
| 540 |
+
"loss": 0.0776,
|
| 541 |
+
"step": 33500
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 7.769652650822669,
|
| 545 |
+
"grad_norm": 0.1155412346124649,
|
| 546 |
+
"learning_rate": 2.883460152182619e-08,
|
| 547 |
+
"loss": 0.0758,
|
| 548 |
+
"step": 34000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 7.883912248628885,
|
| 552 |
+
"grad_norm": 0.0839085578918457,
|
| 553 |
+
"learning_rate": 1.4531723782825104e-08,
|
| 554 |
+
"loss": 0.0762,
|
| 555 |
+
"step": 34500
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 7.998171846435101,
|
| 559 |
+
"grad_norm": 7.536131858825684,
|
| 560 |
+
"learning_rate": 2.2884604382401738e-10,
|
| 561 |
+
"loss": 0.0767,
|
| 562 |
+
"step": 35000
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 8.0,
|
| 566 |
+
"eval_accuracy": 0.9575682256264584,
|
| 567 |
+
"eval_loss": 0.1280793398618698,
|
| 568 |
+
"eval_runtime": 395.9176,
|
| 569 |
+
"eval_samples_per_second": 99.586,
|
| 570 |
+
"eval_steps_per_second": 12.45,
|
| 571 |
+
"step": 35008
|
| 572 |
+
}
|
| 573 |
+
],
|
| 574 |
+
"logging_steps": 500,
|
| 575 |
+
"max_steps": 35008,
|
| 576 |
+
"num_input_tokens_seen": 0,
|
| 577 |
+
"num_train_epochs": 8,
|
| 578 |
+
"save_steps": 500,
|
| 579 |
+
"stateful_callbacks": {
|
| 580 |
+
"TrainerControl": {
|
| 581 |
+
"args": {
|
| 582 |
+
"should_epoch_stop": false,
|
| 583 |
+
"should_evaluate": false,
|
| 584 |
+
"should_log": false,
|
| 585 |
+
"should_save": true,
|
| 586 |
+
"should_training_stop": true
|
| 587 |
+
},
|
| 588 |
+
"attributes": {}
|
| 589 |
+
}
|
| 590 |
+
},
|
| 591 |
+
"total_flos": 8.679226824017687e+19,
|
| 592 |
+
"train_batch_size": 32,
|
| 593 |
+
"trial_name": null,
|
| 594 |
+
"trial_params": null
|
| 595 |
+
}
|
checkpoint-35008/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0aa35f11b4fc0241aa0a1b82ca1a82517a85a325ecb26e9ae7cdd510b8824fab
|
| 3 |
+
size 5240
|
config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vit-base-patch16-224-in21k",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"ViTForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"encoder_stride": 16,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.0,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "Real",
|
| 13 |
+
"1": "Fake"
|
| 14 |
+
},
|
| 15 |
+
"image_size": 224,
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"intermediate_size": 3072,
|
| 18 |
+
"label2id": {
|
| 19 |
+
"Fake": 1,
|
| 20 |
+
"Real": 0
|
| 21 |
+
},
|
| 22 |
+
"layer_norm_eps": 1e-12,
|
| 23 |
+
"model_type": "vit",
|
| 24 |
+
"num_attention_heads": 12,
|
| 25 |
+
"num_channels": 3,
|
| 26 |
+
"num_hidden_layers": 12,
|
| 27 |
+
"patch_size": 16,
|
| 28 |
+
"problem_type": "single_label_classification",
|
| 29 |
+
"qkv_bias": true,
|
| 30 |
+
"torch_dtype": "float32",
|
| 31 |
+
"transformers_version": "4.46.3"
|
| 32 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a3b12bec784c2a5d63adda101dd9a2c52aa4c4409fa27fb3e8030a31c174a17c
|
| 3 |
+
size 343223968
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_normalize": true,
|
| 3 |
+
"do_rescale": true,
|
| 4 |
+
"do_resize": true,
|
| 5 |
+
"image_mean": [
|
| 6 |
+
0.5,
|
| 7 |
+
0.5,
|
| 8 |
+
0.5
|
| 9 |
+
],
|
| 10 |
+
"image_processor_type": "ViTImageProcessor",
|
| 11 |
+
"image_std": [
|
| 12 |
+
0.5,
|
| 13 |
+
0.5,
|
| 14 |
+
0.5
|
| 15 |
+
],
|
| 16 |
+
"resample": 2,
|
| 17 |
+
"rescale_factor": 0.00392156862745098,
|
| 18 |
+
"size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
}
|
| 22 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0aa35f11b4fc0241aa0a1b82ca1a82517a85a325ecb26e9ae7cdd510b8824fab
|
| 3 |
+
size 5240
|