Upload folder using huggingface_hub
Browse files- checkpoint-26970/config.json +45 -0
- checkpoint-26970/model.safetensors +3 -0
- checkpoint-26970/optimizer.pt +3 -0
- checkpoint-26970/preprocessor_config.json +23 -0
- checkpoint-26970/rng_state.pth +3 -0
- checkpoint-26970/scheduler.pt +3 -0
- checkpoint-26970/trainer_state.json +525 -0
- checkpoint-26970/training_args.bin +3 -0
- config.json +45 -0
- model.safetensors +3 -0
- preprocessor_config.json +23 -0
- training_args.bin +3 -0
checkpoint-26970/config.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"ViTForImageClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.0,
|
| 6 |
+
"dtype": "float32",
|
| 7 |
+
"encoder_stride": 16,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.0,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "sad",
|
| 13 |
+
"1": "disgust",
|
| 14 |
+
"2": "angry",
|
| 15 |
+
"3": "neutral",
|
| 16 |
+
"4": "fear",
|
| 17 |
+
"5": "surprise",
|
| 18 |
+
"6": "happy",
|
| 19 |
+
"7": "stress"
|
| 20 |
+
},
|
| 21 |
+
"image_size": 224,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 3072,
|
| 24 |
+
"label2id": {
|
| 25 |
+
"angry": 2,
|
| 26 |
+
"disgust": 1,
|
| 27 |
+
"fear": 4,
|
| 28 |
+
"happy": 6,
|
| 29 |
+
"neutral": 3,
|
| 30 |
+
"sad": 0,
|
| 31 |
+
"stress": 7,
|
| 32 |
+
"surprise": 5
|
| 33 |
+
},
|
| 34 |
+
"layer_norm_eps": 1e-12,
|
| 35 |
+
"model_type": "vit",
|
| 36 |
+
"num_attention_heads": 12,
|
| 37 |
+
"num_channels": 3,
|
| 38 |
+
"num_hidden_layers": 12,
|
| 39 |
+
"patch_size": 16,
|
| 40 |
+
"pooler_act": "tanh",
|
| 41 |
+
"pooler_output_size": 768,
|
| 42 |
+
"problem_type": "single_label_classification",
|
| 43 |
+
"qkv_bias": true,
|
| 44 |
+
"transformers_version": "4.56.1"
|
| 45 |
+
}
|
checkpoint-26970/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ded1dfe872e3a6b927a3b1278573845926edb6bb5b56a07ec6fc570fe532ae6
|
| 3 |
+
size 343242432
|
checkpoint-26970/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b21c4750a94c5261b79446064fc7e84637a670ce09b976169d616bd356cf2d50
|
| 3 |
+
size 52397
|
checkpoint-26970/preprocessor_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": null,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.5,
|
| 8 |
+
0.5,
|
| 9 |
+
0.5
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "ViTImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.5,
|
| 14 |
+
0.5,
|
| 15 |
+
0.5
|
| 16 |
+
],
|
| 17 |
+
"resample": 2,
|
| 18 |
+
"rescale_factor": 0.00392156862745098,
|
| 19 |
+
"size": {
|
| 20 |
+
"height": 224,
|
| 21 |
+
"width": 224
|
| 22 |
+
}
|
| 23 |
+
}
|
checkpoint-26970/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3bf7d5e883080ad6a781cc91583e650a115a8fd8ae23422b7dbeafdf2b3958c
|
| 3 |
+
size 14455
|
checkpoint-26970/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f67fda81f879ae1daee335c91cd2e96cafcf17c550afb35a70e53aba311bde45
|
| 3 |
+
size 1465
|
checkpoint-26970/trainer_state.json
ADDED
|
@@ -0,0 +1,525 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 26970,
|
| 3 |
+
"best_metric": 0.7815828323364258,
|
| 4 |
+
"best_model_checkpoint": "emotion_stress_model_balanced/checkpoint-26970",
|
| 5 |
+
"epoch": 15.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 26970,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.27808676307007785,
|
| 14 |
+
"grad_norm": 1.7513720989227295,
|
| 15 |
+
"learning_rate": 2.949962852897474e-06,
|
| 16 |
+
"loss": 2.0324,
|
| 17 |
+
"step": 500
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.5561735261401557,
|
| 21 |
+
"grad_norm": 1.5362915992736816,
|
| 22 |
+
"learning_rate": 2.89424219910847e-06,
|
| 23 |
+
"loss": 1.9445,
|
| 24 |
+
"step": 1000
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.8342602892102335,
|
| 28 |
+
"grad_norm": 1.602758526802063,
|
| 29 |
+
"learning_rate": 2.838521545319465e-06,
|
| 30 |
+
"loss": 1.864,
|
| 31 |
+
"step": 1500
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 1.0,
|
| 35 |
+
"eval_loss": 1.7139734029769897,
|
| 36 |
+
"eval_runtime": 46.8601,
|
| 37 |
+
"eval_samples_per_second": 801.705,
|
| 38 |
+
"eval_steps_per_second": 100.213,
|
| 39 |
+
"step": 1798
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.1123470522803114,
|
| 43 |
+
"grad_norm": 1.359454870223999,
|
| 44 |
+
"learning_rate": 2.782800891530461e-06,
|
| 45 |
+
"loss": 1.7878,
|
| 46 |
+
"step": 2000
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.3904338153503892,
|
| 50 |
+
"grad_norm": 1.6392908096313477,
|
| 51 |
+
"learning_rate": 2.727080237741456e-06,
|
| 52 |
+
"loss": 1.7208,
|
| 53 |
+
"step": 2500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.668520578420467,
|
| 57 |
+
"grad_norm": 1.5447863340377808,
|
| 58 |
+
"learning_rate": 2.6713595839524517e-06,
|
| 59 |
+
"loss": 1.6523,
|
| 60 |
+
"step": 3000
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.946607341490545,
|
| 64 |
+
"grad_norm": 1.3621774911880493,
|
| 65 |
+
"learning_rate": 2.6156389301634473e-06,
|
| 66 |
+
"loss": 1.5958,
|
| 67 |
+
"step": 3500
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.0,
|
| 71 |
+
"eval_loss": 1.4455991983413696,
|
| 72 |
+
"eval_runtime": 21.1788,
|
| 73 |
+
"eval_samples_per_second": 1773.853,
|
| 74 |
+
"eval_steps_per_second": 221.732,
|
| 75 |
+
"step": 3596
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 2.2246941045606228,
|
| 79 |
+
"grad_norm": 1.2821935415267944,
|
| 80 |
+
"learning_rate": 2.559918276374443e-06,
|
| 81 |
+
"loss": 1.5441,
|
| 82 |
+
"step": 4000
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 2.5027808676307006,
|
| 86 |
+
"grad_norm": 1.2539781332015991,
|
| 87 |
+
"learning_rate": 2.5041976225854384e-06,
|
| 88 |
+
"loss": 1.4902,
|
| 89 |
+
"step": 4500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 2.7808676307007785,
|
| 93 |
+
"grad_norm": 1.247044324874878,
|
| 94 |
+
"learning_rate": 2.448476968796434e-06,
|
| 95 |
+
"loss": 1.4496,
|
| 96 |
+
"step": 5000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.0,
|
| 100 |
+
"eval_loss": 1.2488250732421875,
|
| 101 |
+
"eval_runtime": 21.3725,
|
| 102 |
+
"eval_samples_per_second": 1757.774,
|
| 103 |
+
"eval_steps_per_second": 219.722,
|
| 104 |
+
"step": 5394
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 3.0589543937708563,
|
| 108 |
+
"grad_norm": 1.122979998588562,
|
| 109 |
+
"learning_rate": 2.3927563150074296e-06,
|
| 110 |
+
"loss": 1.4087,
|
| 111 |
+
"step": 5500
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 3.337041156840934,
|
| 115 |
+
"grad_norm": 1.197558045387268,
|
| 116 |
+
"learning_rate": 2.337035661218425e-06,
|
| 117 |
+
"loss": 1.3679,
|
| 118 |
+
"step": 6000
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 3.6151279199110125,
|
| 122 |
+
"grad_norm": 1.2191081047058105,
|
| 123 |
+
"learning_rate": 2.2813150074294203e-06,
|
| 124 |
+
"loss": 1.3373,
|
| 125 |
+
"step": 6500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 3.89321468298109,
|
| 129 |
+
"grad_norm": 1.0837751626968384,
|
| 130 |
+
"learning_rate": 2.2255943536404163e-06,
|
| 131 |
+
"loss": 1.3151,
|
| 132 |
+
"step": 7000
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 4.0,
|
| 136 |
+
"eval_loss": 1.1083296537399292,
|
| 137 |
+
"eval_runtime": 21.3785,
|
| 138 |
+
"eval_samples_per_second": 1757.28,
|
| 139 |
+
"eval_steps_per_second": 219.66,
|
| 140 |
+
"step": 7192
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 4.171301446051168,
|
| 144 |
+
"grad_norm": 1.2327854633331299,
|
| 145 |
+
"learning_rate": 2.1698736998514114e-06,
|
| 146 |
+
"loss": 1.2756,
|
| 147 |
+
"step": 7500
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 4.4493882091212456,
|
| 151 |
+
"grad_norm": 1.0573621988296509,
|
| 152 |
+
"learning_rate": 2.1141530460624074e-06,
|
| 153 |
+
"loss": 1.2522,
|
| 154 |
+
"step": 8000
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 4.727474972191324,
|
| 158 |
+
"grad_norm": 1.0322808027267456,
|
| 159 |
+
"learning_rate": 2.0584323922734026e-06,
|
| 160 |
+
"loss": 1.2318,
|
| 161 |
+
"step": 8500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 5.0,
|
| 165 |
+
"eval_loss": 1.0092095136642456,
|
| 166 |
+
"eval_runtime": 21.2558,
|
| 167 |
+
"eval_samples_per_second": 1767.427,
|
| 168 |
+
"eval_steps_per_second": 220.928,
|
| 169 |
+
"step": 8990
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 5.005561735261401,
|
| 173 |
+
"grad_norm": 0.9742701649665833,
|
| 174 |
+
"learning_rate": 2.002711738484398e-06,
|
| 175 |
+
"loss": 1.2109,
|
| 176 |
+
"step": 9000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 5.2836484983314795,
|
| 180 |
+
"grad_norm": 0.923886239528656,
|
| 181 |
+
"learning_rate": 1.9469910846953937e-06,
|
| 182 |
+
"loss": 1.1863,
|
| 183 |
+
"step": 9500
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 5.561735261401557,
|
| 187 |
+
"grad_norm": 0.8879143595695496,
|
| 188 |
+
"learning_rate": 1.8912704309063893e-06,
|
| 189 |
+
"loss": 1.1806,
|
| 190 |
+
"step": 10000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 5.839822024471635,
|
| 194 |
+
"grad_norm": 1.0213676691055298,
|
| 195 |
+
"learning_rate": 1.835549777117385e-06,
|
| 196 |
+
"loss": 1.1754,
|
| 197 |
+
"step": 10500
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 6.0,
|
| 201 |
+
"eval_loss": 0.9406551122665405,
|
| 202 |
+
"eval_runtime": 21.4277,
|
| 203 |
+
"eval_samples_per_second": 1753.245,
|
| 204 |
+
"eval_steps_per_second": 219.156,
|
| 205 |
+
"step": 10788
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 6.117908787541713,
|
| 209 |
+
"grad_norm": 1.061405897140503,
|
| 210 |
+
"learning_rate": 1.7798291233283805e-06,
|
| 211 |
+
"loss": 1.1517,
|
| 212 |
+
"step": 11000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 6.395995550611791,
|
| 216 |
+
"grad_norm": 0.8349337577819824,
|
| 217 |
+
"learning_rate": 1.7241084695393758e-06,
|
| 218 |
+
"loss": 1.1448,
|
| 219 |
+
"step": 11500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 6.674082313681868,
|
| 223 |
+
"grad_norm": 1.023738980293274,
|
| 224 |
+
"learning_rate": 1.6683878157503716e-06,
|
| 225 |
+
"loss": 1.1415,
|
| 226 |
+
"step": 12000
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 6.952169076751947,
|
| 230 |
+
"grad_norm": 0.9870001673698425,
|
| 231 |
+
"learning_rate": 1.612667161961367e-06,
|
| 232 |
+
"loss": 1.1213,
|
| 233 |
+
"step": 12500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 7.0,
|
| 237 |
+
"eval_loss": 0.8925299048423767,
|
| 238 |
+
"eval_runtime": 21.5166,
|
| 239 |
+
"eval_samples_per_second": 1745.999,
|
| 240 |
+
"eval_steps_per_second": 218.25,
|
| 241 |
+
"step": 12586
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 7.230255839822025,
|
| 245 |
+
"grad_norm": 1.0453287363052368,
|
| 246 |
+
"learning_rate": 1.5569465081723627e-06,
|
| 247 |
+
"loss": 1.1173,
|
| 248 |
+
"step": 13000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 7.508342602892102,
|
| 252 |
+
"grad_norm": 0.8402040004730225,
|
| 253 |
+
"learning_rate": 1.5012258543833581e-06,
|
| 254 |
+
"loss": 1.1053,
|
| 255 |
+
"step": 13500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 7.78642936596218,
|
| 259 |
+
"grad_norm": 0.9418850541114807,
|
| 260 |
+
"learning_rate": 1.4455052005943537e-06,
|
| 261 |
+
"loss": 1.1068,
|
| 262 |
+
"step": 14000
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 8.0,
|
| 266 |
+
"eval_loss": 0.8576251268386841,
|
| 267 |
+
"eval_runtime": 21.4862,
|
| 268 |
+
"eval_samples_per_second": 1748.467,
|
| 269 |
+
"eval_steps_per_second": 218.558,
|
| 270 |
+
"step": 14384
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 8.064516129032258,
|
| 274 |
+
"grad_norm": 0.9174266457557678,
|
| 275 |
+
"learning_rate": 1.3897845468053493e-06,
|
| 276 |
+
"loss": 1.0912,
|
| 277 |
+
"step": 14500
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 8.342602892102336,
|
| 281 |
+
"grad_norm": 0.8031694889068604,
|
| 282 |
+
"learning_rate": 1.3340638930163446e-06,
|
| 283 |
+
"loss": 1.0819,
|
| 284 |
+
"step": 15000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 8.620689655172415,
|
| 288 |
+
"grad_norm": 0.8880692720413208,
|
| 289 |
+
"learning_rate": 1.2783432392273402e-06,
|
| 290 |
+
"loss": 1.0889,
|
| 291 |
+
"step": 15500
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 8.898776418242491,
|
| 295 |
+
"grad_norm": 0.8549273014068604,
|
| 296 |
+
"learning_rate": 1.2226225854383358e-06,
|
| 297 |
+
"loss": 1.0765,
|
| 298 |
+
"step": 16000
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 9.0,
|
| 302 |
+
"eval_loss": 0.8325429558753967,
|
| 303 |
+
"eval_runtime": 21.4809,
|
| 304 |
+
"eval_samples_per_second": 1748.903,
|
| 305 |
+
"eval_steps_per_second": 218.613,
|
| 306 |
+
"step": 16182
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 9.17686318131257,
|
| 310 |
+
"grad_norm": 0.7732747197151184,
|
| 311 |
+
"learning_rate": 1.1669019316493313e-06,
|
| 312 |
+
"loss": 1.0792,
|
| 313 |
+
"step": 16500
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 9.454949944382648,
|
| 317 |
+
"grad_norm": 0.8062688112258911,
|
| 318 |
+
"learning_rate": 1.111181277860327e-06,
|
| 319 |
+
"loss": 1.0736,
|
| 320 |
+
"step": 17000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 9.733036707452726,
|
| 324 |
+
"grad_norm": 0.873096227645874,
|
| 325 |
+
"learning_rate": 1.0554606240713225e-06,
|
| 326 |
+
"loss": 1.0707,
|
| 327 |
+
"step": 17500
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 10.0,
|
| 331 |
+
"eval_loss": 0.8145096302032471,
|
| 332 |
+
"eval_runtime": 21.1956,
|
| 333 |
+
"eval_samples_per_second": 1772.441,
|
| 334 |
+
"eval_steps_per_second": 221.555,
|
| 335 |
+
"step": 17980
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 10.011123470522802,
|
| 339 |
+
"grad_norm": 0.7561053037643433,
|
| 340 |
+
"learning_rate": 9.997399702823179e-07,
|
| 341 |
+
"loss": 1.0761,
|
| 342 |
+
"step": 18000
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 10.28921023359288,
|
| 346 |
+
"grad_norm": 0.770944356918335,
|
| 347 |
+
"learning_rate": 9.440193164933135e-07,
|
| 348 |
+
"loss": 1.0601,
|
| 349 |
+
"step": 18500
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 10.567296996662959,
|
| 353 |
+
"grad_norm": 0.8369573354721069,
|
| 354 |
+
"learning_rate": 8.882986627043091e-07,
|
| 355 |
+
"loss": 1.0777,
|
| 356 |
+
"step": 19000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 10.845383759733037,
|
| 360 |
+
"grad_norm": 0.9367430210113525,
|
| 361 |
+
"learning_rate": 8.325780089153047e-07,
|
| 362 |
+
"loss": 1.0635,
|
| 363 |
+
"step": 19500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 11.0,
|
| 367 |
+
"eval_loss": 0.8015358448028564,
|
| 368 |
+
"eval_runtime": 21.488,
|
| 369 |
+
"eval_samples_per_second": 1748.322,
|
| 370 |
+
"eval_steps_per_second": 218.54,
|
| 371 |
+
"step": 19778
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 11.123470522803114,
|
| 375 |
+
"grad_norm": 0.765078604221344,
|
| 376 |
+
"learning_rate": 7.768573551263003e-07,
|
| 377 |
+
"loss": 1.0579,
|
| 378 |
+
"step": 20000
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 11.401557285873192,
|
| 382 |
+
"grad_norm": 0.7660267949104309,
|
| 383 |
+
"learning_rate": 7.211367013372957e-07,
|
| 384 |
+
"loss": 1.0598,
|
| 385 |
+
"step": 20500
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 11.67964404894327,
|
| 389 |
+
"grad_norm": 0.8136441111564636,
|
| 390 |
+
"learning_rate": 6.654160475482912e-07,
|
| 391 |
+
"loss": 1.0427,
|
| 392 |
+
"step": 21000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 11.957730812013349,
|
| 396 |
+
"grad_norm": 0.7501266598701477,
|
| 397 |
+
"learning_rate": 6.096953937592868e-07,
|
| 398 |
+
"loss": 1.0626,
|
| 399 |
+
"step": 21500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 12.0,
|
| 403 |
+
"eval_loss": 0.7923984527587891,
|
| 404 |
+
"eval_runtime": 21.5573,
|
| 405 |
+
"eval_samples_per_second": 1742.708,
|
| 406 |
+
"eval_steps_per_second": 217.839,
|
| 407 |
+
"step": 21576
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 12.235817575083425,
|
| 411 |
+
"grad_norm": 0.895750105381012,
|
| 412 |
+
"learning_rate": 5.539747399702823e-07,
|
| 413 |
+
"loss": 1.054,
|
| 414 |
+
"step": 22000
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 12.513904338153504,
|
| 418 |
+
"grad_norm": 0.9563328623771667,
|
| 419 |
+
"learning_rate": 4.982540861812778e-07,
|
| 420 |
+
"loss": 1.0396,
|
| 421 |
+
"step": 22500
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 12.791991101223582,
|
| 425 |
+
"grad_norm": 0.8842599391937256,
|
| 426 |
+
"learning_rate": 4.4253343239227344e-07,
|
| 427 |
+
"loss": 1.0475,
|
| 428 |
+
"step": 23000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 13.0,
|
| 432 |
+
"eval_loss": 0.786158561706543,
|
| 433 |
+
"eval_runtime": 21.3873,
|
| 434 |
+
"eval_samples_per_second": 1756.555,
|
| 435 |
+
"eval_steps_per_second": 219.569,
|
| 436 |
+
"step": 23374
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 13.07007786429366,
|
| 440 |
+
"grad_norm": 0.6965476870536804,
|
| 441 |
+
"learning_rate": 3.86812778603269e-07,
|
| 442 |
+
"loss": 1.0429,
|
| 443 |
+
"step": 23500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 13.348164627363737,
|
| 447 |
+
"grad_norm": 0.9398712515830994,
|
| 448 |
+
"learning_rate": 3.310921248142645e-07,
|
| 449 |
+
"loss": 1.0404,
|
| 450 |
+
"step": 24000
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 13.626251390433815,
|
| 454 |
+
"grad_norm": 0.9795758128166199,
|
| 455 |
+
"learning_rate": 2.7537147102526e-07,
|
| 456 |
+
"loss": 1.0469,
|
| 457 |
+
"step": 24500
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 13.904338153503893,
|
| 461 |
+
"grad_norm": 0.8730781674385071,
|
| 462 |
+
"learning_rate": 2.1965081723625558e-07,
|
| 463 |
+
"loss": 1.0361,
|
| 464 |
+
"step": 25000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 14.0,
|
| 468 |
+
"eval_loss": 0.7827068567276001,
|
| 469 |
+
"eval_runtime": 21.4998,
|
| 470 |
+
"eval_samples_per_second": 1747.363,
|
| 471 |
+
"eval_steps_per_second": 218.42,
|
| 472 |
+
"step": 25172
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 14.182424916573972,
|
| 476 |
+
"grad_norm": 0.9993116855621338,
|
| 477 |
+
"learning_rate": 1.6393016344725113e-07,
|
| 478 |
+
"loss": 1.0525,
|
| 479 |
+
"step": 25500
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 14.46051167964405,
|
| 483 |
+
"grad_norm": 0.9638302326202393,
|
| 484 |
+
"learning_rate": 1.0820950965824665e-07,
|
| 485 |
+
"loss": 1.05,
|
| 486 |
+
"step": 26000
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 14.738598442714126,
|
| 490 |
+
"grad_norm": 0.8064911961555481,
|
| 491 |
+
"learning_rate": 5.24888558692422e-08,
|
| 492 |
+
"loss": 1.0457,
|
| 493 |
+
"step": 26500
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 15.0,
|
| 497 |
+
"eval_loss": 0.7815828323364258,
|
| 498 |
+
"eval_runtime": 21.1724,
|
| 499 |
+
"eval_samples_per_second": 1774.387,
|
| 500 |
+
"eval_steps_per_second": 221.798,
|
| 501 |
+
"step": 26970
|
| 502 |
+
}
|
| 503 |
+
],
|
| 504 |
+
"logging_steps": 500,
|
| 505 |
+
"max_steps": 26970,
|
| 506 |
+
"num_input_tokens_seen": 0,
|
| 507 |
+
"num_train_epochs": 15,
|
| 508 |
+
"save_steps": 500,
|
| 509 |
+
"stateful_callbacks": {
|
| 510 |
+
"TrainerControl": {
|
| 511 |
+
"args": {
|
| 512 |
+
"should_epoch_stop": false,
|
| 513 |
+
"should_evaluate": false,
|
| 514 |
+
"should_log": false,
|
| 515 |
+
"should_save": true,
|
| 516 |
+
"should_training_stop": true
|
| 517 |
+
},
|
| 518 |
+
"attributes": {}
|
| 519 |
+
}
|
| 520 |
+
},
|
| 521 |
+
"total_flos": 6.687298365324853e+19,
|
| 522 |
+
"train_batch_size": 32,
|
| 523 |
+
"trial_name": null,
|
| 524 |
+
"trial_params": null
|
| 525 |
+
}
|
checkpoint-26970/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c04666d65140c5836baebf201924f36d50baf00ca3a53018d5a9e91d75e1d6be
|
| 3 |
+
size 5777
|
config.json
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"ViTForImageClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.0,
|
| 6 |
+
"dtype": "float32",
|
| 7 |
+
"encoder_stride": 16,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.0,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "sad",
|
| 13 |
+
"1": "disgust",
|
| 14 |
+
"2": "angry",
|
| 15 |
+
"3": "neutral",
|
| 16 |
+
"4": "fear",
|
| 17 |
+
"5": "surprise",
|
| 18 |
+
"6": "happy",
|
| 19 |
+
"7": "stress"
|
| 20 |
+
},
|
| 21 |
+
"image_size": 224,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 3072,
|
| 24 |
+
"label2id": {
|
| 25 |
+
"angry": 2,
|
| 26 |
+
"disgust": 1,
|
| 27 |
+
"fear": 4,
|
| 28 |
+
"happy": 6,
|
| 29 |
+
"neutral": 3,
|
| 30 |
+
"sad": 0,
|
| 31 |
+
"stress": 7,
|
| 32 |
+
"surprise": 5
|
| 33 |
+
},
|
| 34 |
+
"layer_norm_eps": 1e-12,
|
| 35 |
+
"model_type": "vit",
|
| 36 |
+
"num_attention_heads": 12,
|
| 37 |
+
"num_channels": 3,
|
| 38 |
+
"num_hidden_layers": 12,
|
| 39 |
+
"patch_size": 16,
|
| 40 |
+
"pooler_act": "tanh",
|
| 41 |
+
"pooler_output_size": 768,
|
| 42 |
+
"problem_type": "single_label_classification",
|
| 43 |
+
"qkv_bias": true,
|
| 44 |
+
"transformers_version": "4.56.1"
|
| 45 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ded1dfe872e3a6b927a3b1278573845926edb6bb5b56a07ec6fc570fe532ae6
|
| 3 |
+
size 343242432
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": null,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.5,
|
| 8 |
+
0.5,
|
| 9 |
+
0.5
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "ViTImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.5,
|
| 14 |
+
0.5,
|
| 15 |
+
0.5
|
| 16 |
+
],
|
| 17 |
+
"resample": 2,
|
| 18 |
+
"rescale_factor": 0.00392156862745098,
|
| 19 |
+
"size": {
|
| 20 |
+
"height": 224,
|
| 21 |
+
"width": 224
|
| 22 |
+
}
|
| 23 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c04666d65140c5836baebf201924f36d50baf00ca3a53018d5a9e91d75e1d6be
|
| 3 |
+
size 5777
|