End of training
Browse files- README.md +1 -1
- all_results.json +11 -11
- eval_results.json +6 -6
- runs/May05_19-43-16_503a5f38d136/events.out.tfevents.1714938665.503a5f38d136.8635.3 +3 -0
- train_results.json +6 -6
- trainer_state.json +239 -27
README.md
CHANGED
|
@@ -32,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 0.
|
| 36 |
- Accuracy: 0.8333
|
| 37 |
|
| 38 |
## Model description
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 0.7059
|
| 36 |
- Accuracy: 0.8333
|
| 37 |
|
| 38 |
## Model description
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime": 0.
|
| 6 |
-
"eval_samples_per_second": 12.
|
| 7 |
-
"eval_steps_per_second": 2.
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"eval_accuracy": 0.8333333333333334,
|
| 4 |
+
"eval_loss": 0.7058526873588562,
|
| 5 |
+
"eval_runtime": 0.4657,
|
| 6 |
+
"eval_samples_per_second": 12.883,
|
| 7 |
+
"eval_steps_per_second": 2.147,
|
| 8 |
+
"total_flos": 2.98313540001792e+16,
|
| 9 |
+
"train_loss": 0.40966928482055665,
|
| 10 |
+
"train_runtime": 140.371,
|
| 11 |
+
"train_samples_per_second": 8.549,
|
| 12 |
+
"train_steps_per_second": 0.178
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime": 0.
|
| 6 |
-
"eval_samples_per_second": 12.
|
| 7 |
-
"eval_steps_per_second": 2.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"eval_accuracy": 0.8333333333333334,
|
| 4 |
+
"eval_loss": 0.7058526873588562,
|
| 5 |
+
"eval_runtime": 0.4657,
|
| 6 |
+
"eval_samples_per_second": 12.883,
|
| 7 |
+
"eval_steps_per_second": 2.147
|
| 8 |
}
|
runs/May05_19-43-16_503a5f38d136/events.out.tfevents.1714938665.503a5f38d136.8635.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c5e0130602a07177b8feae7c3cf8f391ec80ff564ee7b0f47e1adf7beae0ddf1
|
| 3 |
+
size 405
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 25.0,
|
| 3 |
+
"total_flos": 2.98313540001792e+16,
|
| 4 |
+
"train_loss": 0.40966928482055665,
|
| 5 |
+
"train_runtime": 140.371,
|
| 6 |
+
"train_samples_per_second": 8.549,
|
| 7 |
+
"train_steps_per_second": 0.178
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,56 +1,268 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-fish/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 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": 1.0,
|
| 13 |
-
"eval_accuracy": 0.
|
| 14 |
-
"eval_loss": 1.
|
| 15 |
-
"eval_runtime":
|
| 16 |
-
"eval_samples_per_second":
|
| 17 |
-
"eval_steps_per_second":
|
| 18 |
"step": 1
|
| 19 |
},
|
| 20 |
{
|
| 21 |
"epoch": 2.0,
|
| 22 |
"eval_accuracy": 0.5,
|
| 23 |
-
"eval_loss": 1.
|
| 24 |
-
"eval_runtime": 0.
|
| 25 |
-
"eval_samples_per_second":
|
| 26 |
-
"eval_steps_per_second":
|
| 27 |
"step": 2
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 3.0,
|
| 31 |
"eval_accuracy": 0.5,
|
| 32 |
-
"eval_loss": 1.
|
| 33 |
-
"eval_runtime": 0.
|
| 34 |
-
"eval_samples_per_second":
|
| 35 |
-
"eval_steps_per_second":
|
| 36 |
"step": 3
|
| 37 |
},
|
| 38 |
{
|
| 39 |
-
"epoch":
|
| 40 |
-
"
|
| 41 |
-
"
|
| 42 |
-
"
|
| 43 |
-
"
|
| 44 |
-
"
|
| 45 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 46 |
}
|
| 47 |
],
|
| 48 |
"logging_steps": 10,
|
| 49 |
-
"max_steps":
|
| 50 |
"num_input_tokens_seen": 0,
|
| 51 |
-
"num_train_epochs":
|
| 52 |
"save_steps": 500,
|
| 53 |
-
"total_flos":
|
| 54 |
"train_batch_size": 32,
|
| 55 |
"trial_name": null,
|
| 56 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8333333333333334,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-fish/checkpoint-13",
|
| 4 |
+
"epoch": 25.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 25,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 1.0,
|
| 13 |
+
"eval_accuracy": 0.5,
|
| 14 |
+
"eval_loss": 1.8802646398544312,
|
| 15 |
+
"eval_runtime": 0.6371,
|
| 16 |
+
"eval_samples_per_second": 9.418,
|
| 17 |
+
"eval_steps_per_second": 1.57,
|
| 18 |
"step": 1
|
| 19 |
},
|
| 20 |
{
|
| 21 |
"epoch": 2.0,
|
| 22 |
"eval_accuracy": 0.5,
|
| 23 |
+
"eval_loss": 1.843906283378601,
|
| 24 |
+
"eval_runtime": 0.6436,
|
| 25 |
+
"eval_samples_per_second": 9.322,
|
| 26 |
+
"eval_steps_per_second": 1.554,
|
| 27 |
"step": 2
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 3.0,
|
| 31 |
"eval_accuracy": 0.5,
|
| 32 |
+
"eval_loss": 1.7571700811386108,
|
| 33 |
+
"eval_runtime": 0.4478,
|
| 34 |
+
"eval_samples_per_second": 13.398,
|
| 35 |
+
"eval_steps_per_second": 2.233,
|
| 36 |
"step": 3
|
| 37 |
},
|
| 38 |
{
|
| 39 |
+
"epoch": 4.0,
|
| 40 |
+
"eval_accuracy": 0.5,
|
| 41 |
+
"eval_loss": 1.625635027885437,
|
| 42 |
+
"eval_runtime": 0.4177,
|
| 43 |
+
"eval_samples_per_second": 14.363,
|
| 44 |
+
"eval_steps_per_second": 2.394,
|
| 45 |
+
"step": 4
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 5.0,
|
| 49 |
+
"eval_accuracy": 0.5,
|
| 50 |
+
"eval_loss": 1.5081933736801147,
|
| 51 |
+
"eval_runtime": 0.43,
|
| 52 |
+
"eval_samples_per_second": 13.952,
|
| 53 |
+
"eval_steps_per_second": 2.325,
|
| 54 |
+
"step": 5
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 6.0,
|
| 58 |
+
"eval_accuracy": 0.5,
|
| 59 |
+
"eval_loss": 1.4301046133041382,
|
| 60 |
+
"eval_runtime": 0.4297,
|
| 61 |
+
"eval_samples_per_second": 13.963,
|
| 62 |
+
"eval_steps_per_second": 2.327,
|
| 63 |
+
"step": 6
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 7.0,
|
| 67 |
+
"eval_accuracy": 0.5,
|
| 68 |
+
"eval_loss": 1.3379350900650024,
|
| 69 |
+
"eval_runtime": 0.6346,
|
| 70 |
+
"eval_samples_per_second": 9.455,
|
| 71 |
+
"eval_steps_per_second": 1.576,
|
| 72 |
+
"step": 7
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 8.0,
|
| 76 |
+
"eval_accuracy": 0.5,
|
| 77 |
+
"eval_loss": 1.226043701171875,
|
| 78 |
+
"eval_runtime": 0.4425,
|
| 79 |
+
"eval_samples_per_second": 13.56,
|
| 80 |
+
"eval_steps_per_second": 2.26,
|
| 81 |
+
"step": 8
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 9.0,
|
| 85 |
+
"eval_accuracy": 0.6666666666666666,
|
| 86 |
+
"eval_loss": 1.1071115732192993,
|
| 87 |
+
"eval_runtime": 0.457,
|
| 88 |
+
"eval_samples_per_second": 13.128,
|
| 89 |
+
"eval_steps_per_second": 2.188,
|
| 90 |
+
"step": 9
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 10.0,
|
| 94 |
+
"grad_norm": 4.3437418937683105,
|
| 95 |
+
"learning_rate": 3.409090909090909e-05,
|
| 96 |
+
"loss": 0.6539,
|
| 97 |
+
"step": 10
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 10.0,
|
| 101 |
+
"eval_accuracy": 0.6666666666666666,
|
| 102 |
+
"eval_loss": 0.9941108822822571,
|
| 103 |
+
"eval_runtime": 0.434,
|
| 104 |
+
"eval_samples_per_second": 13.825,
|
| 105 |
+
"eval_steps_per_second": 2.304,
|
| 106 |
+
"step": 10
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 11.0,
|
| 110 |
+
"eval_accuracy": 0.6666666666666666,
|
| 111 |
+
"eval_loss": 0.8836026191711426,
|
| 112 |
+
"eval_runtime": 0.4206,
|
| 113 |
+
"eval_samples_per_second": 14.264,
|
| 114 |
+
"eval_steps_per_second": 2.377,
|
| 115 |
+
"step": 11
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 12.0,
|
| 119 |
+
"eval_accuracy": 0.6666666666666666,
|
| 120 |
+
"eval_loss": 0.7859137654304504,
|
| 121 |
+
"eval_runtime": 0.6517,
|
| 122 |
+
"eval_samples_per_second": 9.206,
|
| 123 |
+
"eval_steps_per_second": 1.534,
|
| 124 |
+
"step": 12
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 13.0,
|
| 128 |
+
"eval_accuracy": 0.8333333333333334,
|
| 129 |
+
"eval_loss": 0.7058526873588562,
|
| 130 |
+
"eval_runtime": 0.442,
|
| 131 |
+
"eval_samples_per_second": 13.573,
|
| 132 |
+
"eval_steps_per_second": 2.262,
|
| 133 |
+
"step": 13
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 14.0,
|
| 137 |
+
"eval_accuracy": 0.8333333333333334,
|
| 138 |
+
"eval_loss": 0.6357579827308655,
|
| 139 |
+
"eval_runtime": 0.4178,
|
| 140 |
+
"eval_samples_per_second": 14.361,
|
| 141 |
+
"eval_steps_per_second": 2.394,
|
| 142 |
+
"step": 14
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 15.0,
|
| 146 |
+
"eval_accuracy": 0.8333333333333334,
|
| 147 |
+
"eval_loss": 0.5752262473106384,
|
| 148 |
+
"eval_runtime": 0.4294,
|
| 149 |
+
"eval_samples_per_second": 13.972,
|
| 150 |
+
"eval_steps_per_second": 2.329,
|
| 151 |
+
"step": 15
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 16.0,
|
| 155 |
+
"eval_accuracy": 0.8333333333333334,
|
| 156 |
+
"eval_loss": 0.5343325734138489,
|
| 157 |
+
"eval_runtime": 0.4454,
|
| 158 |
+
"eval_samples_per_second": 13.471,
|
| 159 |
+
"eval_steps_per_second": 2.245,
|
| 160 |
+
"step": 16
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 17.0,
|
| 164 |
+
"eval_accuracy": 0.8333333333333334,
|
| 165 |
+
"eval_loss": 0.49941006302833557,
|
| 166 |
+
"eval_runtime": 0.6534,
|
| 167 |
+
"eval_samples_per_second": 9.182,
|
| 168 |
+
"eval_steps_per_second": 1.53,
|
| 169 |
+
"step": 17
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 18.0,
|
| 173 |
+
"eval_accuracy": 0.8333333333333334,
|
| 174 |
+
"eval_loss": 0.47546982765197754,
|
| 175 |
+
"eval_runtime": 0.4476,
|
| 176 |
+
"eval_samples_per_second": 13.404,
|
| 177 |
+
"eval_steps_per_second": 2.234,
|
| 178 |
+
"step": 18
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 19.0,
|
| 182 |
+
"eval_accuracy": 0.8333333333333334,
|
| 183 |
+
"eval_loss": 0.4544316232204437,
|
| 184 |
+
"eval_runtime": 0.4503,
|
| 185 |
+
"eval_samples_per_second": 13.326,
|
| 186 |
+
"eval_steps_per_second": 2.221,
|
| 187 |
+
"step": 19
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 20.0,
|
| 191 |
+
"grad_norm": 4.215967655181885,
|
| 192 |
+
"learning_rate": 1.1363636363636365e-05,
|
| 193 |
+
"loss": 0.2777,
|
| 194 |
+
"step": 20
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 20.0,
|
| 198 |
+
"eval_accuracy": 0.8333333333333334,
|
| 199 |
+
"eval_loss": 0.43280744552612305,
|
| 200 |
+
"eval_runtime": 0.4355,
|
| 201 |
+
"eval_samples_per_second": 13.776,
|
| 202 |
+
"eval_steps_per_second": 2.296,
|
| 203 |
+
"step": 20
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 21.0,
|
| 207 |
+
"eval_accuracy": 0.8333333333333334,
|
| 208 |
+
"eval_loss": 0.41707566380500793,
|
| 209 |
+
"eval_runtime": 0.4926,
|
| 210 |
+
"eval_samples_per_second": 12.18,
|
| 211 |
+
"eval_steps_per_second": 2.03,
|
| 212 |
+
"step": 21
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 22.0,
|
| 216 |
+
"eval_accuracy": 0.8333333333333334,
|
| 217 |
+
"eval_loss": 0.4065686762332916,
|
| 218 |
+
"eval_runtime": 0.6351,
|
| 219 |
+
"eval_samples_per_second": 9.447,
|
| 220 |
+
"eval_steps_per_second": 1.575,
|
| 221 |
+
"step": 22
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 23.0,
|
| 225 |
+
"eval_accuracy": 0.8333333333333334,
|
| 226 |
+
"eval_loss": 0.39954474568367004,
|
| 227 |
+
"eval_runtime": 0.4453,
|
| 228 |
+
"eval_samples_per_second": 13.474,
|
| 229 |
+
"eval_steps_per_second": 2.246,
|
| 230 |
+
"step": 23
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 24.0,
|
| 234 |
+
"eval_accuracy": 0.8333333333333334,
|
| 235 |
+
"eval_loss": 0.39544954895973206,
|
| 236 |
+
"eval_runtime": 0.4307,
|
| 237 |
+
"eval_samples_per_second": 13.93,
|
| 238 |
+
"eval_steps_per_second": 2.322,
|
| 239 |
+
"step": 24
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 25.0,
|
| 243 |
+
"eval_accuracy": 0.8333333333333334,
|
| 244 |
+
"eval_loss": 0.3932792842388153,
|
| 245 |
+
"eval_runtime": 0.4285,
|
| 246 |
+
"eval_samples_per_second": 14.002,
|
| 247 |
+
"eval_steps_per_second": 2.334,
|
| 248 |
+
"step": 25
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 25.0,
|
| 252 |
+
"step": 25,
|
| 253 |
+
"total_flos": 2.98313540001792e+16,
|
| 254 |
+
"train_loss": 0.40966928482055665,
|
| 255 |
+
"train_runtime": 140.371,
|
| 256 |
+
"train_samples_per_second": 8.549,
|
| 257 |
+
"train_steps_per_second": 0.178
|
| 258 |
}
|
| 259 |
],
|
| 260 |
"logging_steps": 10,
|
| 261 |
+
"max_steps": 25,
|
| 262 |
"num_input_tokens_seen": 0,
|
| 263 |
+
"num_train_epochs": 25,
|
| 264 |
"save_steps": 500,
|
| 265 |
+
"total_flos": 2.98313540001792e+16,
|
| 266 |
"train_batch_size": 32,
|
| 267 |
"trial_name": null,
|
| 268 |
"trial_params": null
|