Upload folder using huggingface_hub
#1
by
satwikapaul
- opened
- all_results.json +8 -0
- checkpoint-600/config.json +86 -0
- checkpoint-600/optimizer.pt +3 -0
- checkpoint-600/preprocessor_config.json +22 -0
- checkpoint-600/pytorch_model.bin +3 -0
- checkpoint-600/rng_state.pth +3 -0
- checkpoint-600/scheduler.pt +3 -0
- checkpoint-600/trainer_state.json +511 -0
- checkpoint-600/training_args.bin +3 -0
- checkpoint-640/config.json +86 -0
- checkpoint-640/optimizer.pt +3 -0
- checkpoint-640/preprocessor_config.json +22 -0
- checkpoint-640/pytorch_model.bin +3 -0
- checkpoint-640/rng_state.pth +3 -0
- checkpoint-640/scheduler.pt +3 -0
- checkpoint-640/trainer_state.json +544 -0
- checkpoint-640/training_args.bin +3 -0
- config.json +86 -0
- preprocessor_config.json +22 -0
- pytorch_model.bin +3 -0
- train_results.json +8 -0
- trainer_state.json +553 -0
- training_args.bin +3 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"total_flos": 1.1843641391623373e+18,
|
| 4 |
+
"train_loss": 0.6315841894596815,
|
| 5 |
+
"train_runtime": 473.5366,
|
| 6 |
+
"train_samples_per_second": 32.268,
|
| 7 |
+
"train_steps_per_second": 1.352
|
| 8 |
+
}
|
checkpoint-600/config.json
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "a",
|
| 13 |
+
"1": "b",
|
| 14 |
+
"10": "j",
|
| 15 |
+
"11": "k",
|
| 16 |
+
"12": "l",
|
| 17 |
+
"13": "m",
|
| 18 |
+
"14": "n",
|
| 19 |
+
"15": "o",
|
| 20 |
+
"16": "p",
|
| 21 |
+
"17": "period",
|
| 22 |
+
"18": "q",
|
| 23 |
+
"19": "question%20mark",
|
| 24 |
+
"2": "c",
|
| 25 |
+
"20": "r",
|
| 26 |
+
"21": "s",
|
| 27 |
+
"22": "t",
|
| 28 |
+
"23": "u",
|
| 29 |
+
"24": "v",
|
| 30 |
+
"25": "w",
|
| 31 |
+
"26": "x",
|
| 32 |
+
"27": "y",
|
| 33 |
+
"28": "z",
|
| 34 |
+
"3": "capital",
|
| 35 |
+
"4": "d",
|
| 36 |
+
"5": "e",
|
| 37 |
+
"6": "f",
|
| 38 |
+
"7": "g",
|
| 39 |
+
"8": "h",
|
| 40 |
+
"9": "i"
|
| 41 |
+
},
|
| 42 |
+
"image_size": 224,
|
| 43 |
+
"initializer_range": 0.02,
|
| 44 |
+
"intermediate_size": 3072,
|
| 45 |
+
"label2id": {
|
| 46 |
+
"a": "0",
|
| 47 |
+
"b": "1",
|
| 48 |
+
"c": "2",
|
| 49 |
+
"capital": "3",
|
| 50 |
+
"d": "4",
|
| 51 |
+
"e": "5",
|
| 52 |
+
"f": "6",
|
| 53 |
+
"g": "7",
|
| 54 |
+
"h": "8",
|
| 55 |
+
"i": "9",
|
| 56 |
+
"j": "10",
|
| 57 |
+
"k": "11",
|
| 58 |
+
"l": "12",
|
| 59 |
+
"m": "13",
|
| 60 |
+
"n": "14",
|
| 61 |
+
"o": "15",
|
| 62 |
+
"p": "16",
|
| 63 |
+
"period": "17",
|
| 64 |
+
"q": "18",
|
| 65 |
+
"question%20mark": "19",
|
| 66 |
+
"r": "20",
|
| 67 |
+
"s": "21",
|
| 68 |
+
"t": "22",
|
| 69 |
+
"u": "23",
|
| 70 |
+
"v": "24",
|
| 71 |
+
"w": "25",
|
| 72 |
+
"x": "26",
|
| 73 |
+
"y": "27",
|
| 74 |
+
"z": "28"
|
| 75 |
+
},
|
| 76 |
+
"layer_norm_eps": 1e-12,
|
| 77 |
+
"model_type": "vit",
|
| 78 |
+
"num_attention_heads": 12,
|
| 79 |
+
"num_channels": 3,
|
| 80 |
+
"num_hidden_layers": 12,
|
| 81 |
+
"patch_size": 16,
|
| 82 |
+
"problem_type": "single_label_classification",
|
| 83 |
+
"qkv_bias": true,
|
| 84 |
+
"torch_dtype": "float32",
|
| 85 |
+
"transformers_version": "4.30.2"
|
| 86 |
+
}
|
checkpoint-600/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:53c27f65a8b8e3f973f44c8f7a901b0645535146d1ea086bc93096bbede8ca2b
|
| 3 |
+
size 686684933
|
checkpoint-600/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": "ViTFeatureExtractor",
|
| 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-600/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0c93165ce5edf95f1c1cfa9c27411c5283f28dcf3e706a4405e9e676deba2d58
|
| 3 |
+
size 343351725
|
checkpoint-600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4762ca4e860d7e0ff02cdfb0b71f4f951fb1b0017c35580e28b494788bd64933
|
| 3 |
+
size 14575
|
checkpoint-600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2a3a408b28df808ac711fe39d83558d7ce6c645059f2732c291ddeeab1a5c7ee
|
| 3 |
+
size 627
|
checkpoint-600/trainer_state.json
ADDED
|
@@ -0,0 +1,511 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.16766877472400665,
|
| 3 |
+
"best_model_checkpoint": "./vit-base-beans/checkpoint-560",
|
| 4 |
+
"epoch": 18.75,
|
| 5 |
+
"global_step": 600,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.31,
|
| 12 |
+
"learning_rate": 9.84375e-05,
|
| 13 |
+
"loss": 3.3252,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.62,
|
| 18 |
+
"learning_rate": 9.687500000000001e-05,
|
| 19 |
+
"loss": 3.217,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.94,
|
| 24 |
+
"learning_rate": 9.53125e-05,
|
| 25 |
+
"loss": 3.0799,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.25,
|
| 30 |
+
"learning_rate": 9.375e-05,
|
| 31 |
+
"loss": 2.8394,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.25,
|
| 36 |
+
"eval_accuracy": 0.4815668202764977,
|
| 37 |
+
"eval_loss": 2.7347912788391113,
|
| 38 |
+
"eval_runtime": 5.6305,
|
| 39 |
+
"eval_samples_per_second": 77.081,
|
| 40 |
+
"eval_steps_per_second": 9.768,
|
| 41 |
+
"step": 40
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.56,
|
| 45 |
+
"learning_rate": 9.21875e-05,
|
| 46 |
+
"loss": 2.598,
|
| 47 |
+
"step": 50
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.88,
|
| 51 |
+
"learning_rate": 9.062500000000001e-05,
|
| 52 |
+
"loss": 2.417,
|
| 53 |
+
"step": 60
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 2.19,
|
| 57 |
+
"learning_rate": 8.90625e-05,
|
| 58 |
+
"loss": 2.111,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.5,
|
| 63 |
+
"learning_rate": 8.75e-05,
|
| 64 |
+
"loss": 1.9174,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 2.5,
|
| 69 |
+
"eval_accuracy": 0.8963133640552995,
|
| 70 |
+
"eval_loss": 1.8127654790878296,
|
| 71 |
+
"eval_runtime": 5.3595,
|
| 72 |
+
"eval_samples_per_second": 80.978,
|
| 73 |
+
"eval_steps_per_second": 10.262,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.81,
|
| 78 |
+
"learning_rate": 8.593750000000001e-05,
|
| 79 |
+
"loss": 1.666,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 3.12,
|
| 84 |
+
"learning_rate": 8.4375e-05,
|
| 85 |
+
"loss": 1.5159,
|
| 86 |
+
"step": 100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.44,
|
| 90 |
+
"learning_rate": 8.28125e-05,
|
| 91 |
+
"loss": 1.3105,
|
| 92 |
+
"step": 110
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.75,
|
| 96 |
+
"learning_rate": 8.125000000000001e-05,
|
| 97 |
+
"loss": 1.1859,
|
| 98 |
+
"step": 120
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 3.75,
|
| 102 |
+
"eval_accuracy": 0.9470046082949308,
|
| 103 |
+
"eval_loss": 1.1414676904678345,
|
| 104 |
+
"eval_runtime": 5.3582,
|
| 105 |
+
"eval_samples_per_second": 80.997,
|
| 106 |
+
"eval_steps_per_second": 10.265,
|
| 107 |
+
"step": 120
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 4.06,
|
| 111 |
+
"learning_rate": 7.96875e-05,
|
| 112 |
+
"loss": 1.0525,
|
| 113 |
+
"step": 130
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 4.38,
|
| 117 |
+
"learning_rate": 7.8125e-05,
|
| 118 |
+
"loss": 0.9085,
|
| 119 |
+
"step": 140
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 4.69,
|
| 123 |
+
"learning_rate": 7.65625e-05,
|
| 124 |
+
"loss": 0.8207,
|
| 125 |
+
"step": 150
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 5.0,
|
| 129 |
+
"learning_rate": 7.500000000000001e-05,
|
| 130 |
+
"loss": 0.7413,
|
| 131 |
+
"step": 160
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 5.0,
|
| 135 |
+
"eval_accuracy": 0.9654377880184332,
|
| 136 |
+
"eval_loss": 0.7720305919647217,
|
| 137 |
+
"eval_runtime": 5.6625,
|
| 138 |
+
"eval_samples_per_second": 76.645,
|
| 139 |
+
"eval_steps_per_second": 9.713,
|
| 140 |
+
"step": 160
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 5.31,
|
| 144 |
+
"learning_rate": 7.34375e-05,
|
| 145 |
+
"loss": 0.6334,
|
| 146 |
+
"step": 170
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 5.62,
|
| 150 |
+
"learning_rate": 7.1875e-05,
|
| 151 |
+
"loss": 0.6058,
|
| 152 |
+
"step": 180
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 5.94,
|
| 156 |
+
"learning_rate": 7.031250000000001e-05,
|
| 157 |
+
"loss": 0.5319,
|
| 158 |
+
"step": 190
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 6.25,
|
| 162 |
+
"learning_rate": 6.875e-05,
|
| 163 |
+
"loss": 0.4761,
|
| 164 |
+
"step": 200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 6.25,
|
| 168 |
+
"eval_accuracy": 0.9838709677419355,
|
| 169 |
+
"eval_loss": 0.5084273815155029,
|
| 170 |
+
"eval_runtime": 5.4846,
|
| 171 |
+
"eval_samples_per_second": 79.131,
|
| 172 |
+
"eval_steps_per_second": 10.028,
|
| 173 |
+
"step": 200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 6.56,
|
| 177 |
+
"learning_rate": 6.71875e-05,
|
| 178 |
+
"loss": 0.4109,
|
| 179 |
+
"step": 210
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 6.88,
|
| 183 |
+
"learning_rate": 6.562500000000001e-05,
|
| 184 |
+
"loss": 0.3707,
|
| 185 |
+
"step": 220
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 7.19,
|
| 189 |
+
"learning_rate": 6.40625e-05,
|
| 190 |
+
"loss": 0.3536,
|
| 191 |
+
"step": 230
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 7.5,
|
| 195 |
+
"learning_rate": 6.25e-05,
|
| 196 |
+
"loss": 0.3108,
|
| 197 |
+
"step": 240
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 7.5,
|
| 201 |
+
"eval_accuracy": 0.9746543778801844,
|
| 202 |
+
"eval_loss": 0.36055463552474976,
|
| 203 |
+
"eval_runtime": 5.6696,
|
| 204 |
+
"eval_samples_per_second": 76.549,
|
| 205 |
+
"eval_steps_per_second": 9.701,
|
| 206 |
+
"step": 240
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 7.81,
|
| 210 |
+
"learning_rate": 6.0937500000000004e-05,
|
| 211 |
+
"loss": 0.2789,
|
| 212 |
+
"step": 250
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 8.12,
|
| 216 |
+
"learning_rate": 5.9375e-05,
|
| 217 |
+
"loss": 0.2905,
|
| 218 |
+
"step": 260
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 8.44,
|
| 222 |
+
"learning_rate": 5.78125e-05,
|
| 223 |
+
"loss": 0.2462,
|
| 224 |
+
"step": 270
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 8.75,
|
| 228 |
+
"learning_rate": 5.6250000000000005e-05,
|
| 229 |
+
"loss": 0.251,
|
| 230 |
+
"step": 280
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 8.75,
|
| 234 |
+
"eval_accuracy": 0.9769585253456221,
|
| 235 |
+
"eval_loss": 0.2958492040634155,
|
| 236 |
+
"eval_runtime": 7.6402,
|
| 237 |
+
"eval_samples_per_second": 56.805,
|
| 238 |
+
"eval_steps_per_second": 7.199,
|
| 239 |
+
"step": 280
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 9.06,
|
| 243 |
+
"learning_rate": 5.46875e-05,
|
| 244 |
+
"loss": 0.2171,
|
| 245 |
+
"step": 290
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 9.38,
|
| 249 |
+
"learning_rate": 5.3125000000000004e-05,
|
| 250 |
+
"loss": 0.2064,
|
| 251 |
+
"step": 300
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 9.69,
|
| 255 |
+
"learning_rate": 5.15625e-05,
|
| 256 |
+
"loss": 0.2116,
|
| 257 |
+
"step": 310
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 10.0,
|
| 261 |
+
"learning_rate": 5e-05,
|
| 262 |
+
"loss": 0.1896,
|
| 263 |
+
"step": 320
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 10.0,
|
| 267 |
+
"eval_accuracy": 0.9769585253456221,
|
| 268 |
+
"eval_loss": 0.24788345396518707,
|
| 269 |
+
"eval_runtime": 5.2479,
|
| 270 |
+
"eval_samples_per_second": 82.699,
|
| 271 |
+
"eval_steps_per_second": 10.48,
|
| 272 |
+
"step": 320
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 10.31,
|
| 276 |
+
"learning_rate": 4.8437500000000005e-05,
|
| 277 |
+
"loss": 0.1783,
|
| 278 |
+
"step": 330
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 10.62,
|
| 282 |
+
"learning_rate": 4.6875e-05,
|
| 283 |
+
"loss": 0.1859,
|
| 284 |
+
"step": 340
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 10.94,
|
| 288 |
+
"learning_rate": 4.5312500000000004e-05,
|
| 289 |
+
"loss": 0.1705,
|
| 290 |
+
"step": 350
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 11.25,
|
| 294 |
+
"learning_rate": 4.375e-05,
|
| 295 |
+
"loss": 0.1659,
|
| 296 |
+
"step": 360
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 11.25,
|
| 300 |
+
"eval_accuracy": 0.9838709677419355,
|
| 301 |
+
"eval_loss": 0.23752211034297943,
|
| 302 |
+
"eval_runtime": 5.7417,
|
| 303 |
+
"eval_samples_per_second": 75.588,
|
| 304 |
+
"eval_steps_per_second": 9.579,
|
| 305 |
+
"step": 360
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 11.56,
|
| 309 |
+
"learning_rate": 4.21875e-05,
|
| 310 |
+
"loss": 0.1753,
|
| 311 |
+
"step": 370
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 11.88,
|
| 315 |
+
"learning_rate": 4.0625000000000005e-05,
|
| 316 |
+
"loss": 0.1509,
|
| 317 |
+
"step": 380
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 12.19,
|
| 321 |
+
"learning_rate": 3.90625e-05,
|
| 322 |
+
"loss": 0.1447,
|
| 323 |
+
"step": 390
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 12.5,
|
| 327 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 328 |
+
"loss": 0.1401,
|
| 329 |
+
"step": 400
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 12.5,
|
| 333 |
+
"eval_accuracy": 0.9792626728110599,
|
| 334 |
+
"eval_loss": 0.20333679020404816,
|
| 335 |
+
"eval_runtime": 5.2814,
|
| 336 |
+
"eval_samples_per_second": 82.175,
|
| 337 |
+
"eval_steps_per_second": 10.414,
|
| 338 |
+
"step": 400
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 12.81,
|
| 342 |
+
"learning_rate": 3.59375e-05,
|
| 343 |
+
"loss": 0.1508,
|
| 344 |
+
"step": 410
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 13.12,
|
| 348 |
+
"learning_rate": 3.4375e-05,
|
| 349 |
+
"loss": 0.1364,
|
| 350 |
+
"step": 420
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 13.44,
|
| 354 |
+
"learning_rate": 3.2812500000000005e-05,
|
| 355 |
+
"loss": 0.1415,
|
| 356 |
+
"step": 430
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 13.75,
|
| 360 |
+
"learning_rate": 3.125e-05,
|
| 361 |
+
"loss": 0.131,
|
| 362 |
+
"step": 440
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 13.75,
|
| 366 |
+
"eval_accuracy": 0.9792626728110599,
|
| 367 |
+
"eval_loss": 0.19693893194198608,
|
| 368 |
+
"eval_runtime": 5.0279,
|
| 369 |
+
"eval_samples_per_second": 86.318,
|
| 370 |
+
"eval_steps_per_second": 10.939,
|
| 371 |
+
"step": 440
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 14.06,
|
| 375 |
+
"learning_rate": 2.96875e-05,
|
| 376 |
+
"loss": 0.1264,
|
| 377 |
+
"step": 450
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 14.38,
|
| 381 |
+
"learning_rate": 2.8125000000000003e-05,
|
| 382 |
+
"loss": 0.1376,
|
| 383 |
+
"step": 460
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 14.69,
|
| 387 |
+
"learning_rate": 2.6562500000000002e-05,
|
| 388 |
+
"loss": 0.1211,
|
| 389 |
+
"step": 470
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 15.0,
|
| 393 |
+
"learning_rate": 2.5e-05,
|
| 394 |
+
"loss": 0.1162,
|
| 395 |
+
"step": 480
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 15.0,
|
| 399 |
+
"eval_accuracy": 0.9792626728110599,
|
| 400 |
+
"eval_loss": 0.1791529506444931,
|
| 401 |
+
"eval_runtime": 4.2132,
|
| 402 |
+
"eval_samples_per_second": 103.009,
|
| 403 |
+
"eval_steps_per_second": 13.054,
|
| 404 |
+
"step": 480
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 15.31,
|
| 408 |
+
"learning_rate": 2.34375e-05,
|
| 409 |
+
"loss": 0.1285,
|
| 410 |
+
"step": 490
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 15.62,
|
| 414 |
+
"learning_rate": 2.1875e-05,
|
| 415 |
+
"loss": 0.1136,
|
| 416 |
+
"step": 500
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 15.94,
|
| 420 |
+
"learning_rate": 2.0312500000000002e-05,
|
| 421 |
+
"loss": 0.1117,
|
| 422 |
+
"step": 510
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 16.25,
|
| 426 |
+
"learning_rate": 1.8750000000000002e-05,
|
| 427 |
+
"loss": 0.11,
|
| 428 |
+
"step": 520
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 16.25,
|
| 432 |
+
"eval_accuracy": 0.9792626728110599,
|
| 433 |
+
"eval_loss": 0.17193575203418732,
|
| 434 |
+
"eval_runtime": 4.2988,
|
| 435 |
+
"eval_samples_per_second": 100.957,
|
| 436 |
+
"eval_steps_per_second": 12.794,
|
| 437 |
+
"step": 520
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 16.56,
|
| 441 |
+
"learning_rate": 1.71875e-05,
|
| 442 |
+
"loss": 0.1087,
|
| 443 |
+
"step": 530
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 16.88,
|
| 447 |
+
"learning_rate": 1.5625e-05,
|
| 448 |
+
"loss": 0.1065,
|
| 449 |
+
"step": 540
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 17.19,
|
| 453 |
+
"learning_rate": 1.4062500000000001e-05,
|
| 454 |
+
"loss": 0.1207,
|
| 455 |
+
"step": 550
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 17.5,
|
| 459 |
+
"learning_rate": 1.25e-05,
|
| 460 |
+
"loss": 0.1056,
|
| 461 |
+
"step": 560
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 17.5,
|
| 465 |
+
"eval_accuracy": 0.9792626728110599,
|
| 466 |
+
"eval_loss": 0.16766877472400665,
|
| 467 |
+
"eval_runtime": 5.7239,
|
| 468 |
+
"eval_samples_per_second": 75.823,
|
| 469 |
+
"eval_steps_per_second": 9.609,
|
| 470 |
+
"step": 560
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 17.81,
|
| 474 |
+
"learning_rate": 1.09375e-05,
|
| 475 |
+
"loss": 0.1051,
|
| 476 |
+
"step": 570
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 18.12,
|
| 480 |
+
"learning_rate": 9.375000000000001e-06,
|
| 481 |
+
"loss": 0.1148,
|
| 482 |
+
"step": 580
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 18.44,
|
| 486 |
+
"learning_rate": 7.8125e-06,
|
| 487 |
+
"loss": 0.1043,
|
| 488 |
+
"step": 590
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 18.75,
|
| 492 |
+
"learning_rate": 6.25e-06,
|
| 493 |
+
"loss": 0.1128,
|
| 494 |
+
"step": 600
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 18.75,
|
| 498 |
+
"eval_accuracy": 0.9723502304147466,
|
| 499 |
+
"eval_loss": 0.17274537682533264,
|
| 500 |
+
"eval_runtime": 5.5073,
|
| 501 |
+
"eval_samples_per_second": 78.805,
|
| 502 |
+
"eval_steps_per_second": 9.987,
|
| 503 |
+
"step": 600
|
| 504 |
+
}
|
| 505 |
+
],
|
| 506 |
+
"max_steps": 640,
|
| 507 |
+
"num_train_epochs": 20,
|
| 508 |
+
"total_flos": 1.1105739126909665e+18,
|
| 509 |
+
"trial_name": null,
|
| 510 |
+
"trial_params": null
|
| 511 |
+
}
|
checkpoint-600/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e00db2a0e0945cff474f615b599cebc0ca0436459ecf509b6097643ca635938b
|
| 3 |
+
size 3899
|
checkpoint-640/config.json
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "a",
|
| 13 |
+
"1": "b",
|
| 14 |
+
"10": "j",
|
| 15 |
+
"11": "k",
|
| 16 |
+
"12": "l",
|
| 17 |
+
"13": "m",
|
| 18 |
+
"14": "n",
|
| 19 |
+
"15": "o",
|
| 20 |
+
"16": "p",
|
| 21 |
+
"17": "period",
|
| 22 |
+
"18": "q",
|
| 23 |
+
"19": "question%20mark",
|
| 24 |
+
"2": "c",
|
| 25 |
+
"20": "r",
|
| 26 |
+
"21": "s",
|
| 27 |
+
"22": "t",
|
| 28 |
+
"23": "u",
|
| 29 |
+
"24": "v",
|
| 30 |
+
"25": "w",
|
| 31 |
+
"26": "x",
|
| 32 |
+
"27": "y",
|
| 33 |
+
"28": "z",
|
| 34 |
+
"3": "capital",
|
| 35 |
+
"4": "d",
|
| 36 |
+
"5": "e",
|
| 37 |
+
"6": "f",
|
| 38 |
+
"7": "g",
|
| 39 |
+
"8": "h",
|
| 40 |
+
"9": "i"
|
| 41 |
+
},
|
| 42 |
+
"image_size": 224,
|
| 43 |
+
"initializer_range": 0.02,
|
| 44 |
+
"intermediate_size": 3072,
|
| 45 |
+
"label2id": {
|
| 46 |
+
"a": "0",
|
| 47 |
+
"b": "1",
|
| 48 |
+
"c": "2",
|
| 49 |
+
"capital": "3",
|
| 50 |
+
"d": "4",
|
| 51 |
+
"e": "5",
|
| 52 |
+
"f": "6",
|
| 53 |
+
"g": "7",
|
| 54 |
+
"h": "8",
|
| 55 |
+
"i": "9",
|
| 56 |
+
"j": "10",
|
| 57 |
+
"k": "11",
|
| 58 |
+
"l": "12",
|
| 59 |
+
"m": "13",
|
| 60 |
+
"n": "14",
|
| 61 |
+
"o": "15",
|
| 62 |
+
"p": "16",
|
| 63 |
+
"period": "17",
|
| 64 |
+
"q": "18",
|
| 65 |
+
"question%20mark": "19",
|
| 66 |
+
"r": "20",
|
| 67 |
+
"s": "21",
|
| 68 |
+
"t": "22",
|
| 69 |
+
"u": "23",
|
| 70 |
+
"v": "24",
|
| 71 |
+
"w": "25",
|
| 72 |
+
"x": "26",
|
| 73 |
+
"y": "27",
|
| 74 |
+
"z": "28"
|
| 75 |
+
},
|
| 76 |
+
"layer_norm_eps": 1e-12,
|
| 77 |
+
"model_type": "vit",
|
| 78 |
+
"num_attention_heads": 12,
|
| 79 |
+
"num_channels": 3,
|
| 80 |
+
"num_hidden_layers": 12,
|
| 81 |
+
"patch_size": 16,
|
| 82 |
+
"problem_type": "single_label_classification",
|
| 83 |
+
"qkv_bias": true,
|
| 84 |
+
"torch_dtype": "float32",
|
| 85 |
+
"transformers_version": "4.30.2"
|
| 86 |
+
}
|
checkpoint-640/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a26cdc0df0dea9a25116e0dfff3e12c741831cea206bdc214c326b5d25bee368
|
| 3 |
+
size 686684933
|
checkpoint-640/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": "ViTFeatureExtractor",
|
| 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-640/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:73148133251e579307cedbc6460e628f9c60344cf1f6bcc16883d36df666f0d1
|
| 3 |
+
size 343351725
|
checkpoint-640/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:adb3563d7561e012ae0c950df619e9ba01000208cd5ba07786142cc2783ee2a7
|
| 3 |
+
size 14575
|
checkpoint-640/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc0a2c689052c6bc67cea049e5102d4499d32a5eaf611948ee55d50a010034aa
|
| 3 |
+
size 627
|
checkpoint-640/trainer_state.json
ADDED
|
@@ -0,0 +1,544 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.16443897783756256,
|
| 3 |
+
"best_model_checkpoint": "./vit-base-beans/checkpoint-640",
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"global_step": 640,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.31,
|
| 12 |
+
"learning_rate": 9.84375e-05,
|
| 13 |
+
"loss": 3.3252,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.62,
|
| 18 |
+
"learning_rate": 9.687500000000001e-05,
|
| 19 |
+
"loss": 3.217,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.94,
|
| 24 |
+
"learning_rate": 9.53125e-05,
|
| 25 |
+
"loss": 3.0799,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.25,
|
| 30 |
+
"learning_rate": 9.375e-05,
|
| 31 |
+
"loss": 2.8394,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.25,
|
| 36 |
+
"eval_accuracy": 0.4815668202764977,
|
| 37 |
+
"eval_loss": 2.7347912788391113,
|
| 38 |
+
"eval_runtime": 5.6305,
|
| 39 |
+
"eval_samples_per_second": 77.081,
|
| 40 |
+
"eval_steps_per_second": 9.768,
|
| 41 |
+
"step": 40
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.56,
|
| 45 |
+
"learning_rate": 9.21875e-05,
|
| 46 |
+
"loss": 2.598,
|
| 47 |
+
"step": 50
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.88,
|
| 51 |
+
"learning_rate": 9.062500000000001e-05,
|
| 52 |
+
"loss": 2.417,
|
| 53 |
+
"step": 60
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 2.19,
|
| 57 |
+
"learning_rate": 8.90625e-05,
|
| 58 |
+
"loss": 2.111,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.5,
|
| 63 |
+
"learning_rate": 8.75e-05,
|
| 64 |
+
"loss": 1.9174,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 2.5,
|
| 69 |
+
"eval_accuracy": 0.8963133640552995,
|
| 70 |
+
"eval_loss": 1.8127654790878296,
|
| 71 |
+
"eval_runtime": 5.3595,
|
| 72 |
+
"eval_samples_per_second": 80.978,
|
| 73 |
+
"eval_steps_per_second": 10.262,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.81,
|
| 78 |
+
"learning_rate": 8.593750000000001e-05,
|
| 79 |
+
"loss": 1.666,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 3.12,
|
| 84 |
+
"learning_rate": 8.4375e-05,
|
| 85 |
+
"loss": 1.5159,
|
| 86 |
+
"step": 100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.44,
|
| 90 |
+
"learning_rate": 8.28125e-05,
|
| 91 |
+
"loss": 1.3105,
|
| 92 |
+
"step": 110
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.75,
|
| 96 |
+
"learning_rate": 8.125000000000001e-05,
|
| 97 |
+
"loss": 1.1859,
|
| 98 |
+
"step": 120
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 3.75,
|
| 102 |
+
"eval_accuracy": 0.9470046082949308,
|
| 103 |
+
"eval_loss": 1.1414676904678345,
|
| 104 |
+
"eval_runtime": 5.3582,
|
| 105 |
+
"eval_samples_per_second": 80.997,
|
| 106 |
+
"eval_steps_per_second": 10.265,
|
| 107 |
+
"step": 120
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 4.06,
|
| 111 |
+
"learning_rate": 7.96875e-05,
|
| 112 |
+
"loss": 1.0525,
|
| 113 |
+
"step": 130
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 4.38,
|
| 117 |
+
"learning_rate": 7.8125e-05,
|
| 118 |
+
"loss": 0.9085,
|
| 119 |
+
"step": 140
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 4.69,
|
| 123 |
+
"learning_rate": 7.65625e-05,
|
| 124 |
+
"loss": 0.8207,
|
| 125 |
+
"step": 150
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 5.0,
|
| 129 |
+
"learning_rate": 7.500000000000001e-05,
|
| 130 |
+
"loss": 0.7413,
|
| 131 |
+
"step": 160
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 5.0,
|
| 135 |
+
"eval_accuracy": 0.9654377880184332,
|
| 136 |
+
"eval_loss": 0.7720305919647217,
|
| 137 |
+
"eval_runtime": 5.6625,
|
| 138 |
+
"eval_samples_per_second": 76.645,
|
| 139 |
+
"eval_steps_per_second": 9.713,
|
| 140 |
+
"step": 160
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 5.31,
|
| 144 |
+
"learning_rate": 7.34375e-05,
|
| 145 |
+
"loss": 0.6334,
|
| 146 |
+
"step": 170
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 5.62,
|
| 150 |
+
"learning_rate": 7.1875e-05,
|
| 151 |
+
"loss": 0.6058,
|
| 152 |
+
"step": 180
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 5.94,
|
| 156 |
+
"learning_rate": 7.031250000000001e-05,
|
| 157 |
+
"loss": 0.5319,
|
| 158 |
+
"step": 190
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 6.25,
|
| 162 |
+
"learning_rate": 6.875e-05,
|
| 163 |
+
"loss": 0.4761,
|
| 164 |
+
"step": 200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 6.25,
|
| 168 |
+
"eval_accuracy": 0.9838709677419355,
|
| 169 |
+
"eval_loss": 0.5084273815155029,
|
| 170 |
+
"eval_runtime": 5.4846,
|
| 171 |
+
"eval_samples_per_second": 79.131,
|
| 172 |
+
"eval_steps_per_second": 10.028,
|
| 173 |
+
"step": 200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 6.56,
|
| 177 |
+
"learning_rate": 6.71875e-05,
|
| 178 |
+
"loss": 0.4109,
|
| 179 |
+
"step": 210
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 6.88,
|
| 183 |
+
"learning_rate": 6.562500000000001e-05,
|
| 184 |
+
"loss": 0.3707,
|
| 185 |
+
"step": 220
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 7.19,
|
| 189 |
+
"learning_rate": 6.40625e-05,
|
| 190 |
+
"loss": 0.3536,
|
| 191 |
+
"step": 230
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 7.5,
|
| 195 |
+
"learning_rate": 6.25e-05,
|
| 196 |
+
"loss": 0.3108,
|
| 197 |
+
"step": 240
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 7.5,
|
| 201 |
+
"eval_accuracy": 0.9746543778801844,
|
| 202 |
+
"eval_loss": 0.36055463552474976,
|
| 203 |
+
"eval_runtime": 5.6696,
|
| 204 |
+
"eval_samples_per_second": 76.549,
|
| 205 |
+
"eval_steps_per_second": 9.701,
|
| 206 |
+
"step": 240
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 7.81,
|
| 210 |
+
"learning_rate": 6.0937500000000004e-05,
|
| 211 |
+
"loss": 0.2789,
|
| 212 |
+
"step": 250
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 8.12,
|
| 216 |
+
"learning_rate": 5.9375e-05,
|
| 217 |
+
"loss": 0.2905,
|
| 218 |
+
"step": 260
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 8.44,
|
| 222 |
+
"learning_rate": 5.78125e-05,
|
| 223 |
+
"loss": 0.2462,
|
| 224 |
+
"step": 270
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 8.75,
|
| 228 |
+
"learning_rate": 5.6250000000000005e-05,
|
| 229 |
+
"loss": 0.251,
|
| 230 |
+
"step": 280
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 8.75,
|
| 234 |
+
"eval_accuracy": 0.9769585253456221,
|
| 235 |
+
"eval_loss": 0.2958492040634155,
|
| 236 |
+
"eval_runtime": 7.6402,
|
| 237 |
+
"eval_samples_per_second": 56.805,
|
| 238 |
+
"eval_steps_per_second": 7.199,
|
| 239 |
+
"step": 280
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 9.06,
|
| 243 |
+
"learning_rate": 5.46875e-05,
|
| 244 |
+
"loss": 0.2171,
|
| 245 |
+
"step": 290
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 9.38,
|
| 249 |
+
"learning_rate": 5.3125000000000004e-05,
|
| 250 |
+
"loss": 0.2064,
|
| 251 |
+
"step": 300
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 9.69,
|
| 255 |
+
"learning_rate": 5.15625e-05,
|
| 256 |
+
"loss": 0.2116,
|
| 257 |
+
"step": 310
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 10.0,
|
| 261 |
+
"learning_rate": 5e-05,
|
| 262 |
+
"loss": 0.1896,
|
| 263 |
+
"step": 320
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 10.0,
|
| 267 |
+
"eval_accuracy": 0.9769585253456221,
|
| 268 |
+
"eval_loss": 0.24788345396518707,
|
| 269 |
+
"eval_runtime": 5.2479,
|
| 270 |
+
"eval_samples_per_second": 82.699,
|
| 271 |
+
"eval_steps_per_second": 10.48,
|
| 272 |
+
"step": 320
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 10.31,
|
| 276 |
+
"learning_rate": 4.8437500000000005e-05,
|
| 277 |
+
"loss": 0.1783,
|
| 278 |
+
"step": 330
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 10.62,
|
| 282 |
+
"learning_rate": 4.6875e-05,
|
| 283 |
+
"loss": 0.1859,
|
| 284 |
+
"step": 340
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 10.94,
|
| 288 |
+
"learning_rate": 4.5312500000000004e-05,
|
| 289 |
+
"loss": 0.1705,
|
| 290 |
+
"step": 350
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 11.25,
|
| 294 |
+
"learning_rate": 4.375e-05,
|
| 295 |
+
"loss": 0.1659,
|
| 296 |
+
"step": 360
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 11.25,
|
| 300 |
+
"eval_accuracy": 0.9838709677419355,
|
| 301 |
+
"eval_loss": 0.23752211034297943,
|
| 302 |
+
"eval_runtime": 5.7417,
|
| 303 |
+
"eval_samples_per_second": 75.588,
|
| 304 |
+
"eval_steps_per_second": 9.579,
|
| 305 |
+
"step": 360
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 11.56,
|
| 309 |
+
"learning_rate": 4.21875e-05,
|
| 310 |
+
"loss": 0.1753,
|
| 311 |
+
"step": 370
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 11.88,
|
| 315 |
+
"learning_rate": 4.0625000000000005e-05,
|
| 316 |
+
"loss": 0.1509,
|
| 317 |
+
"step": 380
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 12.19,
|
| 321 |
+
"learning_rate": 3.90625e-05,
|
| 322 |
+
"loss": 0.1447,
|
| 323 |
+
"step": 390
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 12.5,
|
| 327 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 328 |
+
"loss": 0.1401,
|
| 329 |
+
"step": 400
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 12.5,
|
| 333 |
+
"eval_accuracy": 0.9792626728110599,
|
| 334 |
+
"eval_loss": 0.20333679020404816,
|
| 335 |
+
"eval_runtime": 5.2814,
|
| 336 |
+
"eval_samples_per_second": 82.175,
|
| 337 |
+
"eval_steps_per_second": 10.414,
|
| 338 |
+
"step": 400
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 12.81,
|
| 342 |
+
"learning_rate": 3.59375e-05,
|
| 343 |
+
"loss": 0.1508,
|
| 344 |
+
"step": 410
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 13.12,
|
| 348 |
+
"learning_rate": 3.4375e-05,
|
| 349 |
+
"loss": 0.1364,
|
| 350 |
+
"step": 420
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 13.44,
|
| 354 |
+
"learning_rate": 3.2812500000000005e-05,
|
| 355 |
+
"loss": 0.1415,
|
| 356 |
+
"step": 430
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 13.75,
|
| 360 |
+
"learning_rate": 3.125e-05,
|
| 361 |
+
"loss": 0.131,
|
| 362 |
+
"step": 440
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 13.75,
|
| 366 |
+
"eval_accuracy": 0.9792626728110599,
|
| 367 |
+
"eval_loss": 0.19693893194198608,
|
| 368 |
+
"eval_runtime": 5.0279,
|
| 369 |
+
"eval_samples_per_second": 86.318,
|
| 370 |
+
"eval_steps_per_second": 10.939,
|
| 371 |
+
"step": 440
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 14.06,
|
| 375 |
+
"learning_rate": 2.96875e-05,
|
| 376 |
+
"loss": 0.1264,
|
| 377 |
+
"step": 450
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 14.38,
|
| 381 |
+
"learning_rate": 2.8125000000000003e-05,
|
| 382 |
+
"loss": 0.1376,
|
| 383 |
+
"step": 460
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 14.69,
|
| 387 |
+
"learning_rate": 2.6562500000000002e-05,
|
| 388 |
+
"loss": 0.1211,
|
| 389 |
+
"step": 470
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 15.0,
|
| 393 |
+
"learning_rate": 2.5e-05,
|
| 394 |
+
"loss": 0.1162,
|
| 395 |
+
"step": 480
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 15.0,
|
| 399 |
+
"eval_accuracy": 0.9792626728110599,
|
| 400 |
+
"eval_loss": 0.1791529506444931,
|
| 401 |
+
"eval_runtime": 4.2132,
|
| 402 |
+
"eval_samples_per_second": 103.009,
|
| 403 |
+
"eval_steps_per_second": 13.054,
|
| 404 |
+
"step": 480
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 15.31,
|
| 408 |
+
"learning_rate": 2.34375e-05,
|
| 409 |
+
"loss": 0.1285,
|
| 410 |
+
"step": 490
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 15.62,
|
| 414 |
+
"learning_rate": 2.1875e-05,
|
| 415 |
+
"loss": 0.1136,
|
| 416 |
+
"step": 500
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 15.94,
|
| 420 |
+
"learning_rate": 2.0312500000000002e-05,
|
| 421 |
+
"loss": 0.1117,
|
| 422 |
+
"step": 510
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 16.25,
|
| 426 |
+
"learning_rate": 1.8750000000000002e-05,
|
| 427 |
+
"loss": 0.11,
|
| 428 |
+
"step": 520
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 16.25,
|
| 432 |
+
"eval_accuracy": 0.9792626728110599,
|
| 433 |
+
"eval_loss": 0.17193575203418732,
|
| 434 |
+
"eval_runtime": 4.2988,
|
| 435 |
+
"eval_samples_per_second": 100.957,
|
| 436 |
+
"eval_steps_per_second": 12.794,
|
| 437 |
+
"step": 520
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 16.56,
|
| 441 |
+
"learning_rate": 1.71875e-05,
|
| 442 |
+
"loss": 0.1087,
|
| 443 |
+
"step": 530
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 16.88,
|
| 447 |
+
"learning_rate": 1.5625e-05,
|
| 448 |
+
"loss": 0.1065,
|
| 449 |
+
"step": 540
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 17.19,
|
| 453 |
+
"learning_rate": 1.4062500000000001e-05,
|
| 454 |
+
"loss": 0.1207,
|
| 455 |
+
"step": 550
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 17.5,
|
| 459 |
+
"learning_rate": 1.25e-05,
|
| 460 |
+
"loss": 0.1056,
|
| 461 |
+
"step": 560
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 17.5,
|
| 465 |
+
"eval_accuracy": 0.9792626728110599,
|
| 466 |
+
"eval_loss": 0.16766877472400665,
|
| 467 |
+
"eval_runtime": 5.7239,
|
| 468 |
+
"eval_samples_per_second": 75.823,
|
| 469 |
+
"eval_steps_per_second": 9.609,
|
| 470 |
+
"step": 560
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 17.81,
|
| 474 |
+
"learning_rate": 1.09375e-05,
|
| 475 |
+
"loss": 0.1051,
|
| 476 |
+
"step": 570
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 18.12,
|
| 480 |
+
"learning_rate": 9.375000000000001e-06,
|
| 481 |
+
"loss": 0.1148,
|
| 482 |
+
"step": 580
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 18.44,
|
| 486 |
+
"learning_rate": 7.8125e-06,
|
| 487 |
+
"loss": 0.1043,
|
| 488 |
+
"step": 590
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 18.75,
|
| 492 |
+
"learning_rate": 6.25e-06,
|
| 493 |
+
"loss": 0.1128,
|
| 494 |
+
"step": 600
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 18.75,
|
| 498 |
+
"eval_accuracy": 0.9723502304147466,
|
| 499 |
+
"eval_loss": 0.17274537682533264,
|
| 500 |
+
"eval_runtime": 5.5073,
|
| 501 |
+
"eval_samples_per_second": 78.805,
|
| 502 |
+
"eval_steps_per_second": 9.987,
|
| 503 |
+
"step": 600
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 19.06,
|
| 507 |
+
"learning_rate": 4.6875000000000004e-06,
|
| 508 |
+
"loss": 0.102,
|
| 509 |
+
"step": 610
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 19.38,
|
| 513 |
+
"learning_rate": 3.125e-06,
|
| 514 |
+
"loss": 0.1028,
|
| 515 |
+
"step": 620
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 19.69,
|
| 519 |
+
"learning_rate": 1.5625e-06,
|
| 520 |
+
"loss": 0.1096,
|
| 521 |
+
"step": 630
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 20.0,
|
| 525 |
+
"learning_rate": 0.0,
|
| 526 |
+
"loss": 0.1018,
|
| 527 |
+
"step": 640
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 20.0,
|
| 531 |
+
"eval_accuracy": 0.9792626728110599,
|
| 532 |
+
"eval_loss": 0.16443897783756256,
|
| 533 |
+
"eval_runtime": 4.9159,
|
| 534 |
+
"eval_samples_per_second": 88.285,
|
| 535 |
+
"eval_steps_per_second": 11.188,
|
| 536 |
+
"step": 640
|
| 537 |
+
}
|
| 538 |
+
],
|
| 539 |
+
"max_steps": 640,
|
| 540 |
+
"num_train_epochs": 20,
|
| 541 |
+
"total_flos": 1.1843641391623373e+18,
|
| 542 |
+
"trial_name": null,
|
| 543 |
+
"trial_params": null
|
| 544 |
+
}
|
checkpoint-640/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e00db2a0e0945cff474f615b599cebc0ca0436459ecf509b6097643ca635938b
|
| 3 |
+
size 3899
|
config.json
ADDED
|
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": "a",
|
| 13 |
+
"1": "b",
|
| 14 |
+
"10": "j",
|
| 15 |
+
"11": "k",
|
| 16 |
+
"12": "l",
|
| 17 |
+
"13": "m",
|
| 18 |
+
"14": "n",
|
| 19 |
+
"15": "o",
|
| 20 |
+
"16": "p",
|
| 21 |
+
"17": "period",
|
| 22 |
+
"18": "q",
|
| 23 |
+
"19": "question%20mark",
|
| 24 |
+
"2": "c",
|
| 25 |
+
"20": "r",
|
| 26 |
+
"21": "s",
|
| 27 |
+
"22": "t",
|
| 28 |
+
"23": "u",
|
| 29 |
+
"24": "v",
|
| 30 |
+
"25": "w",
|
| 31 |
+
"26": "x",
|
| 32 |
+
"27": "y",
|
| 33 |
+
"28": "z",
|
| 34 |
+
"3": "capital",
|
| 35 |
+
"4": "d",
|
| 36 |
+
"5": "e",
|
| 37 |
+
"6": "f",
|
| 38 |
+
"7": "g",
|
| 39 |
+
"8": "h",
|
| 40 |
+
"9": "i"
|
| 41 |
+
},
|
| 42 |
+
"image_size": 224,
|
| 43 |
+
"initializer_range": 0.02,
|
| 44 |
+
"intermediate_size": 3072,
|
| 45 |
+
"label2id": {
|
| 46 |
+
"a": "0",
|
| 47 |
+
"b": "1",
|
| 48 |
+
"c": "2",
|
| 49 |
+
"capital": "3",
|
| 50 |
+
"d": "4",
|
| 51 |
+
"e": "5",
|
| 52 |
+
"f": "6",
|
| 53 |
+
"g": "7",
|
| 54 |
+
"h": "8",
|
| 55 |
+
"i": "9",
|
| 56 |
+
"j": "10",
|
| 57 |
+
"k": "11",
|
| 58 |
+
"l": "12",
|
| 59 |
+
"m": "13",
|
| 60 |
+
"n": "14",
|
| 61 |
+
"o": "15",
|
| 62 |
+
"p": "16",
|
| 63 |
+
"period": "17",
|
| 64 |
+
"q": "18",
|
| 65 |
+
"question%20mark": "19",
|
| 66 |
+
"r": "20",
|
| 67 |
+
"s": "21",
|
| 68 |
+
"t": "22",
|
| 69 |
+
"u": "23",
|
| 70 |
+
"v": "24",
|
| 71 |
+
"w": "25",
|
| 72 |
+
"x": "26",
|
| 73 |
+
"y": "27",
|
| 74 |
+
"z": "28"
|
| 75 |
+
},
|
| 76 |
+
"layer_norm_eps": 1e-12,
|
| 77 |
+
"model_type": "vit",
|
| 78 |
+
"num_attention_heads": 12,
|
| 79 |
+
"num_channels": 3,
|
| 80 |
+
"num_hidden_layers": 12,
|
| 81 |
+
"patch_size": 16,
|
| 82 |
+
"problem_type": "single_label_classification",
|
| 83 |
+
"qkv_bias": true,
|
| 84 |
+
"torch_dtype": "float32",
|
| 85 |
+
"transformers_version": "4.30.2"
|
| 86 |
+
}
|
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": "ViTFeatureExtractor",
|
| 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 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:73148133251e579307cedbc6460e628f9c60344cf1f6bcc16883d36df666f0d1
|
| 3 |
+
size 343351725
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 20.0,
|
| 3 |
+
"total_flos": 1.1843641391623373e+18,
|
| 4 |
+
"train_loss": 0.6315841894596815,
|
| 5 |
+
"train_runtime": 473.5366,
|
| 6 |
+
"train_samples_per_second": 32.268,
|
| 7 |
+
"train_steps_per_second": 1.352
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,553 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.16443897783756256,
|
| 3 |
+
"best_model_checkpoint": "./vit-base-beans/checkpoint-640",
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"global_step": 640,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.31,
|
| 12 |
+
"learning_rate": 9.84375e-05,
|
| 13 |
+
"loss": 3.3252,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.62,
|
| 18 |
+
"learning_rate": 9.687500000000001e-05,
|
| 19 |
+
"loss": 3.217,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.94,
|
| 24 |
+
"learning_rate": 9.53125e-05,
|
| 25 |
+
"loss": 3.0799,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.25,
|
| 30 |
+
"learning_rate": 9.375e-05,
|
| 31 |
+
"loss": 2.8394,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.25,
|
| 36 |
+
"eval_accuracy": 0.4815668202764977,
|
| 37 |
+
"eval_loss": 2.7347912788391113,
|
| 38 |
+
"eval_runtime": 5.6305,
|
| 39 |
+
"eval_samples_per_second": 77.081,
|
| 40 |
+
"eval_steps_per_second": 9.768,
|
| 41 |
+
"step": 40
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.56,
|
| 45 |
+
"learning_rate": 9.21875e-05,
|
| 46 |
+
"loss": 2.598,
|
| 47 |
+
"step": 50
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.88,
|
| 51 |
+
"learning_rate": 9.062500000000001e-05,
|
| 52 |
+
"loss": 2.417,
|
| 53 |
+
"step": 60
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 2.19,
|
| 57 |
+
"learning_rate": 8.90625e-05,
|
| 58 |
+
"loss": 2.111,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.5,
|
| 63 |
+
"learning_rate": 8.75e-05,
|
| 64 |
+
"loss": 1.9174,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 2.5,
|
| 69 |
+
"eval_accuracy": 0.8963133640552995,
|
| 70 |
+
"eval_loss": 1.8127654790878296,
|
| 71 |
+
"eval_runtime": 5.3595,
|
| 72 |
+
"eval_samples_per_second": 80.978,
|
| 73 |
+
"eval_steps_per_second": 10.262,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.81,
|
| 78 |
+
"learning_rate": 8.593750000000001e-05,
|
| 79 |
+
"loss": 1.666,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 3.12,
|
| 84 |
+
"learning_rate": 8.4375e-05,
|
| 85 |
+
"loss": 1.5159,
|
| 86 |
+
"step": 100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.44,
|
| 90 |
+
"learning_rate": 8.28125e-05,
|
| 91 |
+
"loss": 1.3105,
|
| 92 |
+
"step": 110
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.75,
|
| 96 |
+
"learning_rate": 8.125000000000001e-05,
|
| 97 |
+
"loss": 1.1859,
|
| 98 |
+
"step": 120
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 3.75,
|
| 102 |
+
"eval_accuracy": 0.9470046082949308,
|
| 103 |
+
"eval_loss": 1.1414676904678345,
|
| 104 |
+
"eval_runtime": 5.3582,
|
| 105 |
+
"eval_samples_per_second": 80.997,
|
| 106 |
+
"eval_steps_per_second": 10.265,
|
| 107 |
+
"step": 120
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 4.06,
|
| 111 |
+
"learning_rate": 7.96875e-05,
|
| 112 |
+
"loss": 1.0525,
|
| 113 |
+
"step": 130
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 4.38,
|
| 117 |
+
"learning_rate": 7.8125e-05,
|
| 118 |
+
"loss": 0.9085,
|
| 119 |
+
"step": 140
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 4.69,
|
| 123 |
+
"learning_rate": 7.65625e-05,
|
| 124 |
+
"loss": 0.8207,
|
| 125 |
+
"step": 150
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 5.0,
|
| 129 |
+
"learning_rate": 7.500000000000001e-05,
|
| 130 |
+
"loss": 0.7413,
|
| 131 |
+
"step": 160
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 5.0,
|
| 135 |
+
"eval_accuracy": 0.9654377880184332,
|
| 136 |
+
"eval_loss": 0.7720305919647217,
|
| 137 |
+
"eval_runtime": 5.6625,
|
| 138 |
+
"eval_samples_per_second": 76.645,
|
| 139 |
+
"eval_steps_per_second": 9.713,
|
| 140 |
+
"step": 160
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 5.31,
|
| 144 |
+
"learning_rate": 7.34375e-05,
|
| 145 |
+
"loss": 0.6334,
|
| 146 |
+
"step": 170
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 5.62,
|
| 150 |
+
"learning_rate": 7.1875e-05,
|
| 151 |
+
"loss": 0.6058,
|
| 152 |
+
"step": 180
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 5.94,
|
| 156 |
+
"learning_rate": 7.031250000000001e-05,
|
| 157 |
+
"loss": 0.5319,
|
| 158 |
+
"step": 190
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 6.25,
|
| 162 |
+
"learning_rate": 6.875e-05,
|
| 163 |
+
"loss": 0.4761,
|
| 164 |
+
"step": 200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 6.25,
|
| 168 |
+
"eval_accuracy": 0.9838709677419355,
|
| 169 |
+
"eval_loss": 0.5084273815155029,
|
| 170 |
+
"eval_runtime": 5.4846,
|
| 171 |
+
"eval_samples_per_second": 79.131,
|
| 172 |
+
"eval_steps_per_second": 10.028,
|
| 173 |
+
"step": 200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 6.56,
|
| 177 |
+
"learning_rate": 6.71875e-05,
|
| 178 |
+
"loss": 0.4109,
|
| 179 |
+
"step": 210
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 6.88,
|
| 183 |
+
"learning_rate": 6.562500000000001e-05,
|
| 184 |
+
"loss": 0.3707,
|
| 185 |
+
"step": 220
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 7.19,
|
| 189 |
+
"learning_rate": 6.40625e-05,
|
| 190 |
+
"loss": 0.3536,
|
| 191 |
+
"step": 230
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 7.5,
|
| 195 |
+
"learning_rate": 6.25e-05,
|
| 196 |
+
"loss": 0.3108,
|
| 197 |
+
"step": 240
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 7.5,
|
| 201 |
+
"eval_accuracy": 0.9746543778801844,
|
| 202 |
+
"eval_loss": 0.36055463552474976,
|
| 203 |
+
"eval_runtime": 5.6696,
|
| 204 |
+
"eval_samples_per_second": 76.549,
|
| 205 |
+
"eval_steps_per_second": 9.701,
|
| 206 |
+
"step": 240
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 7.81,
|
| 210 |
+
"learning_rate": 6.0937500000000004e-05,
|
| 211 |
+
"loss": 0.2789,
|
| 212 |
+
"step": 250
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 8.12,
|
| 216 |
+
"learning_rate": 5.9375e-05,
|
| 217 |
+
"loss": 0.2905,
|
| 218 |
+
"step": 260
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 8.44,
|
| 222 |
+
"learning_rate": 5.78125e-05,
|
| 223 |
+
"loss": 0.2462,
|
| 224 |
+
"step": 270
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 8.75,
|
| 228 |
+
"learning_rate": 5.6250000000000005e-05,
|
| 229 |
+
"loss": 0.251,
|
| 230 |
+
"step": 280
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 8.75,
|
| 234 |
+
"eval_accuracy": 0.9769585253456221,
|
| 235 |
+
"eval_loss": 0.2958492040634155,
|
| 236 |
+
"eval_runtime": 7.6402,
|
| 237 |
+
"eval_samples_per_second": 56.805,
|
| 238 |
+
"eval_steps_per_second": 7.199,
|
| 239 |
+
"step": 280
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 9.06,
|
| 243 |
+
"learning_rate": 5.46875e-05,
|
| 244 |
+
"loss": 0.2171,
|
| 245 |
+
"step": 290
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 9.38,
|
| 249 |
+
"learning_rate": 5.3125000000000004e-05,
|
| 250 |
+
"loss": 0.2064,
|
| 251 |
+
"step": 300
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 9.69,
|
| 255 |
+
"learning_rate": 5.15625e-05,
|
| 256 |
+
"loss": 0.2116,
|
| 257 |
+
"step": 310
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 10.0,
|
| 261 |
+
"learning_rate": 5e-05,
|
| 262 |
+
"loss": 0.1896,
|
| 263 |
+
"step": 320
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 10.0,
|
| 267 |
+
"eval_accuracy": 0.9769585253456221,
|
| 268 |
+
"eval_loss": 0.24788345396518707,
|
| 269 |
+
"eval_runtime": 5.2479,
|
| 270 |
+
"eval_samples_per_second": 82.699,
|
| 271 |
+
"eval_steps_per_second": 10.48,
|
| 272 |
+
"step": 320
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 10.31,
|
| 276 |
+
"learning_rate": 4.8437500000000005e-05,
|
| 277 |
+
"loss": 0.1783,
|
| 278 |
+
"step": 330
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 10.62,
|
| 282 |
+
"learning_rate": 4.6875e-05,
|
| 283 |
+
"loss": 0.1859,
|
| 284 |
+
"step": 340
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 10.94,
|
| 288 |
+
"learning_rate": 4.5312500000000004e-05,
|
| 289 |
+
"loss": 0.1705,
|
| 290 |
+
"step": 350
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 11.25,
|
| 294 |
+
"learning_rate": 4.375e-05,
|
| 295 |
+
"loss": 0.1659,
|
| 296 |
+
"step": 360
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 11.25,
|
| 300 |
+
"eval_accuracy": 0.9838709677419355,
|
| 301 |
+
"eval_loss": 0.23752211034297943,
|
| 302 |
+
"eval_runtime": 5.7417,
|
| 303 |
+
"eval_samples_per_second": 75.588,
|
| 304 |
+
"eval_steps_per_second": 9.579,
|
| 305 |
+
"step": 360
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 11.56,
|
| 309 |
+
"learning_rate": 4.21875e-05,
|
| 310 |
+
"loss": 0.1753,
|
| 311 |
+
"step": 370
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 11.88,
|
| 315 |
+
"learning_rate": 4.0625000000000005e-05,
|
| 316 |
+
"loss": 0.1509,
|
| 317 |
+
"step": 380
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 12.19,
|
| 321 |
+
"learning_rate": 3.90625e-05,
|
| 322 |
+
"loss": 0.1447,
|
| 323 |
+
"step": 390
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 12.5,
|
| 327 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 328 |
+
"loss": 0.1401,
|
| 329 |
+
"step": 400
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 12.5,
|
| 333 |
+
"eval_accuracy": 0.9792626728110599,
|
| 334 |
+
"eval_loss": 0.20333679020404816,
|
| 335 |
+
"eval_runtime": 5.2814,
|
| 336 |
+
"eval_samples_per_second": 82.175,
|
| 337 |
+
"eval_steps_per_second": 10.414,
|
| 338 |
+
"step": 400
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 12.81,
|
| 342 |
+
"learning_rate": 3.59375e-05,
|
| 343 |
+
"loss": 0.1508,
|
| 344 |
+
"step": 410
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 13.12,
|
| 348 |
+
"learning_rate": 3.4375e-05,
|
| 349 |
+
"loss": 0.1364,
|
| 350 |
+
"step": 420
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 13.44,
|
| 354 |
+
"learning_rate": 3.2812500000000005e-05,
|
| 355 |
+
"loss": 0.1415,
|
| 356 |
+
"step": 430
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 13.75,
|
| 360 |
+
"learning_rate": 3.125e-05,
|
| 361 |
+
"loss": 0.131,
|
| 362 |
+
"step": 440
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 13.75,
|
| 366 |
+
"eval_accuracy": 0.9792626728110599,
|
| 367 |
+
"eval_loss": 0.19693893194198608,
|
| 368 |
+
"eval_runtime": 5.0279,
|
| 369 |
+
"eval_samples_per_second": 86.318,
|
| 370 |
+
"eval_steps_per_second": 10.939,
|
| 371 |
+
"step": 440
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 14.06,
|
| 375 |
+
"learning_rate": 2.96875e-05,
|
| 376 |
+
"loss": 0.1264,
|
| 377 |
+
"step": 450
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 14.38,
|
| 381 |
+
"learning_rate": 2.8125000000000003e-05,
|
| 382 |
+
"loss": 0.1376,
|
| 383 |
+
"step": 460
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 14.69,
|
| 387 |
+
"learning_rate": 2.6562500000000002e-05,
|
| 388 |
+
"loss": 0.1211,
|
| 389 |
+
"step": 470
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 15.0,
|
| 393 |
+
"learning_rate": 2.5e-05,
|
| 394 |
+
"loss": 0.1162,
|
| 395 |
+
"step": 480
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 15.0,
|
| 399 |
+
"eval_accuracy": 0.9792626728110599,
|
| 400 |
+
"eval_loss": 0.1791529506444931,
|
| 401 |
+
"eval_runtime": 4.2132,
|
| 402 |
+
"eval_samples_per_second": 103.009,
|
| 403 |
+
"eval_steps_per_second": 13.054,
|
| 404 |
+
"step": 480
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 15.31,
|
| 408 |
+
"learning_rate": 2.34375e-05,
|
| 409 |
+
"loss": 0.1285,
|
| 410 |
+
"step": 490
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 15.62,
|
| 414 |
+
"learning_rate": 2.1875e-05,
|
| 415 |
+
"loss": 0.1136,
|
| 416 |
+
"step": 500
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 15.94,
|
| 420 |
+
"learning_rate": 2.0312500000000002e-05,
|
| 421 |
+
"loss": 0.1117,
|
| 422 |
+
"step": 510
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 16.25,
|
| 426 |
+
"learning_rate": 1.8750000000000002e-05,
|
| 427 |
+
"loss": 0.11,
|
| 428 |
+
"step": 520
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 16.25,
|
| 432 |
+
"eval_accuracy": 0.9792626728110599,
|
| 433 |
+
"eval_loss": 0.17193575203418732,
|
| 434 |
+
"eval_runtime": 4.2988,
|
| 435 |
+
"eval_samples_per_second": 100.957,
|
| 436 |
+
"eval_steps_per_second": 12.794,
|
| 437 |
+
"step": 520
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 16.56,
|
| 441 |
+
"learning_rate": 1.71875e-05,
|
| 442 |
+
"loss": 0.1087,
|
| 443 |
+
"step": 530
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 16.88,
|
| 447 |
+
"learning_rate": 1.5625e-05,
|
| 448 |
+
"loss": 0.1065,
|
| 449 |
+
"step": 540
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 17.19,
|
| 453 |
+
"learning_rate": 1.4062500000000001e-05,
|
| 454 |
+
"loss": 0.1207,
|
| 455 |
+
"step": 550
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 17.5,
|
| 459 |
+
"learning_rate": 1.25e-05,
|
| 460 |
+
"loss": 0.1056,
|
| 461 |
+
"step": 560
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 17.5,
|
| 465 |
+
"eval_accuracy": 0.9792626728110599,
|
| 466 |
+
"eval_loss": 0.16766877472400665,
|
| 467 |
+
"eval_runtime": 5.7239,
|
| 468 |
+
"eval_samples_per_second": 75.823,
|
| 469 |
+
"eval_steps_per_second": 9.609,
|
| 470 |
+
"step": 560
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 17.81,
|
| 474 |
+
"learning_rate": 1.09375e-05,
|
| 475 |
+
"loss": 0.1051,
|
| 476 |
+
"step": 570
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 18.12,
|
| 480 |
+
"learning_rate": 9.375000000000001e-06,
|
| 481 |
+
"loss": 0.1148,
|
| 482 |
+
"step": 580
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 18.44,
|
| 486 |
+
"learning_rate": 7.8125e-06,
|
| 487 |
+
"loss": 0.1043,
|
| 488 |
+
"step": 590
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 18.75,
|
| 492 |
+
"learning_rate": 6.25e-06,
|
| 493 |
+
"loss": 0.1128,
|
| 494 |
+
"step": 600
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 18.75,
|
| 498 |
+
"eval_accuracy": 0.9723502304147466,
|
| 499 |
+
"eval_loss": 0.17274537682533264,
|
| 500 |
+
"eval_runtime": 5.5073,
|
| 501 |
+
"eval_samples_per_second": 78.805,
|
| 502 |
+
"eval_steps_per_second": 9.987,
|
| 503 |
+
"step": 600
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 19.06,
|
| 507 |
+
"learning_rate": 4.6875000000000004e-06,
|
| 508 |
+
"loss": 0.102,
|
| 509 |
+
"step": 610
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 19.38,
|
| 513 |
+
"learning_rate": 3.125e-06,
|
| 514 |
+
"loss": 0.1028,
|
| 515 |
+
"step": 620
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 19.69,
|
| 519 |
+
"learning_rate": 1.5625e-06,
|
| 520 |
+
"loss": 0.1096,
|
| 521 |
+
"step": 630
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 20.0,
|
| 525 |
+
"learning_rate": 0.0,
|
| 526 |
+
"loss": 0.1018,
|
| 527 |
+
"step": 640
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 20.0,
|
| 531 |
+
"eval_accuracy": 0.9792626728110599,
|
| 532 |
+
"eval_loss": 0.16443897783756256,
|
| 533 |
+
"eval_runtime": 4.9159,
|
| 534 |
+
"eval_samples_per_second": 88.285,
|
| 535 |
+
"eval_steps_per_second": 11.188,
|
| 536 |
+
"step": 640
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 20.0,
|
| 540 |
+
"step": 640,
|
| 541 |
+
"total_flos": 1.1843641391623373e+18,
|
| 542 |
+
"train_loss": 0.6315841894596815,
|
| 543 |
+
"train_runtime": 473.5366,
|
| 544 |
+
"train_samples_per_second": 32.268,
|
| 545 |
+
"train_steps_per_second": 1.352
|
| 546 |
+
}
|
| 547 |
+
],
|
| 548 |
+
"max_steps": 640,
|
| 549 |
+
"num_train_epochs": 20,
|
| 550 |
+
"total_flos": 1.1843641391623373e+18,
|
| 551 |
+
"trial_name": null,
|
| 552 |
+
"trial_params": null
|
| 553 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e00db2a0e0945cff474f615b599cebc0ca0436459ecf509b6097643ca635938b
|
| 3 |
+
size 3899
|