🍻 cheers
Browse files- README.md +3 -2
- all_results.json +5 -5
- train_results.json +5 -5
- trainer_state.json +251 -26
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: other
|
| 3 |
base_model: apple/mobilevit-xx-small
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
datasets:
|
| 7 |
- imagefolder
|
|
@@ -14,7 +15,7 @@ model-index:
|
|
| 14 |
name: Image Classification
|
| 15 |
type: image-classification
|
| 16 |
dataset:
|
| 17 |
-
name:
|
| 18 |
type: imagefolder
|
| 19 |
config: default
|
| 20 |
split: train
|
|
@@ -30,7 +31,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 30 |
|
| 31 |
# vit-base-plankton
|
| 32 |
|
| 33 |
-
This model is a fine-tuned version of [apple/mobilevit-xx-small](https://huggingface.co/apple/mobilevit-xx-small) on the
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
- Loss: 0.7642
|
| 36 |
- Accuracy: 0.8051
|
|
|
|
| 2 |
license: other
|
| 3 |
base_model: apple/mobilevit-xx-small
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
datasets:
|
| 8 |
- imagefolder
|
|
|
|
| 15 |
name: Image Classification
|
| 16 |
type: image-classification
|
| 17 |
dataset:
|
| 18 |
+
name: plankton_fairscope
|
| 19 |
type: imagefolder
|
| 20 |
config: default
|
| 21 |
split: train
|
|
|
|
| 31 |
|
| 32 |
# vit-base-plankton
|
| 33 |
|
| 34 |
+
This model is a fine-tuned version of [apple/mobilevit-xx-small](https://huggingface.co/apple/mobilevit-xx-small) on the plankton_fairscope dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
- Loss: 0.7642
|
| 37 |
- Accuracy: 0.8051
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
+
"total_flos": 6911768078843904.0,
|
| 4 |
+
"train_loss": 1.2305065219600995,
|
| 5 |
+
"train_runtime": 2179.2388,
|
| 6 |
+
"train_samples_per_second": 2.815,
|
| 7 |
+
"train_steps_per_second": 0.176
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 2.0,
|
| 3 |
+
"total_flos": 6911768078843904.0,
|
| 4 |
+
"train_loss": 1.2305065219600995,
|
| 5 |
+
"train_runtime": 2179.2388,
|
| 6 |
+
"train_samples_per_second": 2.815,
|
| 7 |
+
"train_steps_per_second": 0.176
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,58 +1,283 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint":
|
| 4 |
"epoch": 2.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
-
"epoch": 0.
|
| 13 |
-
"learning_rate": 0.
|
| 14 |
-
"loss":
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"learning_rate": 0.
|
| 20 |
-
"loss":
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
-
"epoch":
|
| 25 |
-
"learning_rate":
|
| 26 |
-
"loss": 2.
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
-
"epoch":
|
| 31 |
-
"learning_rate":
|
| 32 |
-
"loss":
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"epoch":
|
| 37 |
-
"learning_rate":
|
| 38 |
-
"loss":
|
| 39 |
"step": 50
|
| 40 |
},
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 41 |
{
|
| 42 |
"epoch": 2.0,
|
| 43 |
-
"step":
|
| 44 |
-
"total_flos":
|
| 45 |
-
"train_loss":
|
| 46 |
-
"train_runtime":
|
| 47 |
-
"train_samples_per_second":
|
| 48 |
-
"train_steps_per_second": 0.
|
| 49 |
}
|
| 50 |
],
|
| 51 |
"logging_steps": 10,
|
| 52 |
-
"max_steps":
|
| 53 |
"num_train_epochs": 2,
|
| 54 |
"save_steps": 100,
|
| 55 |
-
"total_flos":
|
| 56 |
"trial_name": null,
|
| 57 |
"trial_params": null
|
| 58 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.7642104625701904,
|
| 3 |
+
"best_model_checkpoint": "./vit-base-plankton/checkpoint-300",
|
| 4 |
"epoch": 2.0,
|
| 5 |
"eval_steps": 100,
|
| 6 |
+
"global_step": 384,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.05,
|
| 13 |
+
"learning_rate": 0.00019479166666666668,
|
| 14 |
+
"loss": 2.4716,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.1,
|
| 19 |
+
"learning_rate": 0.00018958333333333332,
|
| 20 |
+
"loss": 2.296,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.16,
|
| 25 |
+
"learning_rate": 0.000184375,
|
| 26 |
+
"loss": 2.1599,
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
+
"epoch": 0.21,
|
| 31 |
+
"learning_rate": 0.0001791666666666667,
|
| 32 |
+
"loss": 1.9804,
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"epoch": 0.26,
|
| 37 |
+
"learning_rate": 0.00017395833333333334,
|
| 38 |
+
"loss": 1.7886,
|
| 39 |
"step": 50
|
| 40 |
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.31,
|
| 43 |
+
"learning_rate": 0.00016875,
|
| 44 |
+
"loss": 1.7123,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.36,
|
| 49 |
+
"learning_rate": 0.00016354166666666668,
|
| 50 |
+
"loss": 1.6042,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.42,
|
| 55 |
+
"learning_rate": 0.00015833333333333332,
|
| 56 |
+
"loss": 1.6366,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.47,
|
| 61 |
+
"learning_rate": 0.000153125,
|
| 62 |
+
"loss": 1.5387,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.52,
|
| 67 |
+
"learning_rate": 0.0001479166666666667,
|
| 68 |
+
"loss": 1.5476,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.52,
|
| 73 |
+
"eval_accuracy": 0.741851368970013,
|
| 74 |
+
"eval_loss": 1.2744733095169067,
|
| 75 |
+
"eval_runtime": 213.9361,
|
| 76 |
+
"eval_samples_per_second": 14.341,
|
| 77 |
+
"eval_steps_per_second": 1.795,
|
| 78 |
+
"step": 100
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.57,
|
| 82 |
+
"learning_rate": 0.00014270833333333334,
|
| 83 |
+
"loss": 1.3298,
|
| 84 |
+
"step": 110
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.62,
|
| 88 |
+
"learning_rate": 0.0001375,
|
| 89 |
+
"loss": 1.19,
|
| 90 |
+
"step": 120
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.68,
|
| 94 |
+
"learning_rate": 0.00013229166666666668,
|
| 95 |
+
"loss": 1.0872,
|
| 96 |
+
"step": 130
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.73,
|
| 100 |
+
"learning_rate": 0.00012708333333333332,
|
| 101 |
+
"loss": 1.1765,
|
| 102 |
+
"step": 140
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.78,
|
| 106 |
+
"learning_rate": 0.00012187500000000001,
|
| 107 |
+
"loss": 1.1879,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.83,
|
| 112 |
+
"learning_rate": 0.00011666666666666668,
|
| 113 |
+
"loss": 1.0871,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.89,
|
| 118 |
+
"learning_rate": 0.00011145833333333334,
|
| 119 |
+
"loss": 1.1028,
|
| 120 |
+
"step": 170
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.94,
|
| 124 |
+
"learning_rate": 0.00010625000000000001,
|
| 125 |
+
"loss": 1.1018,
|
| 126 |
+
"step": 180
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.99,
|
| 130 |
+
"learning_rate": 0.00010104166666666668,
|
| 131 |
+
"loss": 1.1769,
|
| 132 |
+
"step": 190
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 1.04,
|
| 136 |
+
"learning_rate": 9.583333333333334e-05,
|
| 137 |
+
"loss": 1.0997,
|
| 138 |
+
"step": 200
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 1.04,
|
| 142 |
+
"eval_accuracy": 0.7842242503259452,
|
| 143 |
+
"eval_loss": 0.8653025031089783,
|
| 144 |
+
"eval_runtime": 205.776,
|
| 145 |
+
"eval_samples_per_second": 14.909,
|
| 146 |
+
"eval_steps_per_second": 1.866,
|
| 147 |
+
"step": 200
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.09,
|
| 151 |
+
"learning_rate": 9.062500000000001e-05,
|
| 152 |
+
"loss": 0.9956,
|
| 153 |
+
"step": 210
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.15,
|
| 157 |
+
"learning_rate": 8.541666666666666e-05,
|
| 158 |
+
"loss": 1.0083,
|
| 159 |
+
"step": 220
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.2,
|
| 163 |
+
"learning_rate": 8.020833333333334e-05,
|
| 164 |
+
"loss": 0.9768,
|
| 165 |
+
"step": 230
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.25,
|
| 169 |
+
"learning_rate": 7.500000000000001e-05,
|
| 170 |
+
"loss": 0.965,
|
| 171 |
+
"step": 240
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.3,
|
| 175 |
+
"learning_rate": 6.979166666666666e-05,
|
| 176 |
+
"loss": 0.9682,
|
| 177 |
+
"step": 250
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.35,
|
| 181 |
+
"learning_rate": 6.458333333333334e-05,
|
| 182 |
+
"loss": 0.9361,
|
| 183 |
+
"step": 260
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.41,
|
| 187 |
+
"learning_rate": 5.9375e-05,
|
| 188 |
+
"loss": 0.9819,
|
| 189 |
+
"step": 270
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 1.46,
|
| 193 |
+
"learning_rate": 5.4166666666666664e-05,
|
| 194 |
+
"loss": 0.8521,
|
| 195 |
+
"step": 280
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.51,
|
| 199 |
+
"learning_rate": 4.8958333333333335e-05,
|
| 200 |
+
"loss": 0.9011,
|
| 201 |
+
"step": 290
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.56,
|
| 205 |
+
"learning_rate": 4.375e-05,
|
| 206 |
+
"loss": 0.9498,
|
| 207 |
+
"step": 300
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 1.56,
|
| 211 |
+
"eval_accuracy": 0.8050847457627118,
|
| 212 |
+
"eval_loss": 0.7642104625701904,
|
| 213 |
+
"eval_runtime": 203.4405,
|
| 214 |
+
"eval_samples_per_second": 15.081,
|
| 215 |
+
"eval_steps_per_second": 1.888,
|
| 216 |
+
"step": 300
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 1.61,
|
| 220 |
+
"learning_rate": 3.854166666666667e-05,
|
| 221 |
+
"loss": 0.9927,
|
| 222 |
+
"step": 310
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 1.67,
|
| 226 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 227 |
+
"loss": 0.8678,
|
| 228 |
+
"step": 320
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.72,
|
| 232 |
+
"learning_rate": 2.8125000000000003e-05,
|
| 233 |
+
"loss": 0.8756,
|
| 234 |
+
"step": 330
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.77,
|
| 238 |
+
"learning_rate": 2.2916666666666667e-05,
|
| 239 |
+
"loss": 0.8625,
|
| 240 |
+
"step": 340
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.82,
|
| 244 |
+
"learning_rate": 1.7708333333333335e-05,
|
| 245 |
+
"loss": 0.9351,
|
| 246 |
+
"step": 350
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 1.88,
|
| 250 |
+
"learning_rate": 1.25e-05,
|
| 251 |
+
"loss": 0.9369,
|
| 252 |
+
"step": 360
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.93,
|
| 256 |
+
"learning_rate": 7.2916666666666674e-06,
|
| 257 |
+
"loss": 0.9611,
|
| 258 |
+
"step": 370
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 1.98,
|
| 262 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 263 |
+
"loss": 0.7286,
|
| 264 |
+
"step": 380
|
| 265 |
+
},
|
| 266 |
{
|
| 267 |
"epoch": 2.0,
|
| 268 |
+
"step": 384,
|
| 269 |
+
"total_flos": 6911768078843904.0,
|
| 270 |
+
"train_loss": 1.2305065219600995,
|
| 271 |
+
"train_runtime": 2179.2388,
|
| 272 |
+
"train_samples_per_second": 2.815,
|
| 273 |
+
"train_steps_per_second": 0.176
|
| 274 |
}
|
| 275 |
],
|
| 276 |
"logging_steps": 10,
|
| 277 |
+
"max_steps": 384,
|
| 278 |
"num_train_epochs": 2,
|
| 279 |
"save_steps": 100,
|
| 280 |
+
"total_flos": 6911768078843904.0,
|
| 281 |
"trial_name": null,
|
| 282 |
"trial_params": null
|
| 283 |
}
|