model 6 output
Browse files- checkpoint-40/config.json +82 -0
- checkpoint-40/optimizer.pt +3 -0
- checkpoint-40/pytorch_model.bin +3 -0
- checkpoint-40/rng_state.pth +3 -0
- checkpoint-40/scheduler.pt +3 -0
- checkpoint-40/trainer_state.json +302 -0
- checkpoint-40/training_args.bin +3 -0
- checkpoint-60/config.json +82 -0
- checkpoint-60/optimizer.pt +3 -0
- checkpoint-60/pytorch_model.bin +3 -0
- checkpoint-60/rng_state.pth +3 -0
- checkpoint-60/scheduler.pt +3 -0
- checkpoint-60/trainer_state.json +445 -0
- checkpoint-60/training_args.bin +3 -0
- checkpoint-80/config.json +82 -0
- checkpoint-80/optimizer.pt +3 -0
- checkpoint-80/pytorch_model.bin +3 -0
- checkpoint-80/rng_state.pth +3 -0
- checkpoint-80/scheduler.pt +3 -0
- checkpoint-80/trainer_state.json +588 -0
- checkpoint-80/training_args.bin +3 -0
- config.json +82 -0
- pytorch_model.bin +3 -0
- runs/Jul14_06-17-20_a2366e527b16/1657779466.1818385/events.out.tfevents.1657779466.a2366e527b16.71.1 +3 -0
- runs/Jul14_06-17-20_a2366e527b16/events.out.tfevents.1657779466.a2366e527b16.71.0 +3 -0
- training_args.bin +3 -0
checkpoint-40/config.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "nvidia/mit-b0",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SegformerForSemanticSegmentation"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"classifier_dropout_prob": 0.1,
|
| 8 |
+
"decoder_hidden_size": 256,
|
| 9 |
+
"depths": [
|
| 10 |
+
2,
|
| 11 |
+
2,
|
| 12 |
+
2,
|
| 13 |
+
2
|
| 14 |
+
],
|
| 15 |
+
"downsampling_rates": [
|
| 16 |
+
1,
|
| 17 |
+
4,
|
| 18 |
+
8,
|
| 19 |
+
16
|
| 20 |
+
],
|
| 21 |
+
"drop_path_rate": 0.1,
|
| 22 |
+
"hidden_act": "gelu",
|
| 23 |
+
"hidden_dropout_prob": 0.0,
|
| 24 |
+
"hidden_sizes": [
|
| 25 |
+
32,
|
| 26 |
+
64,
|
| 27 |
+
160,
|
| 28 |
+
256
|
| 29 |
+
],
|
| 30 |
+
"id2label": {
|
| 31 |
+
"0": "Nothing",
|
| 32 |
+
"1": "Sand",
|
| 33 |
+
"2": "Soil",
|
| 34 |
+
"3": "Rock"
|
| 35 |
+
},
|
| 36 |
+
"image_size": 224,
|
| 37 |
+
"initializer_range": 0.02,
|
| 38 |
+
"label2id": {
|
| 39 |
+
"Nothing": 0,
|
| 40 |
+
"Rock": 3,
|
| 41 |
+
"Sand": 1,
|
| 42 |
+
"Soil": 2
|
| 43 |
+
},
|
| 44 |
+
"layer_norm_eps": 1e-06,
|
| 45 |
+
"mlp_ratios": [
|
| 46 |
+
4,
|
| 47 |
+
4,
|
| 48 |
+
4,
|
| 49 |
+
4
|
| 50 |
+
],
|
| 51 |
+
"model_type": "segformer",
|
| 52 |
+
"num_attention_heads": [
|
| 53 |
+
1,
|
| 54 |
+
2,
|
| 55 |
+
5,
|
| 56 |
+
8
|
| 57 |
+
],
|
| 58 |
+
"num_channels": 3,
|
| 59 |
+
"num_encoder_blocks": 4,
|
| 60 |
+
"patch_sizes": [
|
| 61 |
+
7,
|
| 62 |
+
3,
|
| 63 |
+
3,
|
| 64 |
+
3
|
| 65 |
+
],
|
| 66 |
+
"reshape_last_stage": true,
|
| 67 |
+
"semantic_loss_ignore_index": 255,
|
| 68 |
+
"sr_ratios": [
|
| 69 |
+
8,
|
| 70 |
+
4,
|
| 71 |
+
2,
|
| 72 |
+
1
|
| 73 |
+
],
|
| 74 |
+
"strides": [
|
| 75 |
+
4,
|
| 76 |
+
2,
|
| 77 |
+
2,
|
| 78 |
+
2
|
| 79 |
+
],
|
| 80 |
+
"torch_dtype": "float32",
|
| 81 |
+
"transformers_version": "4.20.1"
|
| 82 |
+
}
|
checkpoint-40/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f29442d97dd8854f875dcd54f760388895a1bb37c43027afd5f54e7c94b60113
|
| 3 |
+
size 29839501
|
checkpoint-40/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e3b00e2121ebe4691e8b8f8019da3e89926dbb9a2e13c9c862f3009fad240b9
|
| 3 |
+
size 14932001
|
checkpoint-40/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2e22e87e4e2ce800709dddb89169eda44f4fe5cb3c15df905a7daa50aa35869
|
| 3 |
+
size 14503
|
checkpoint-40/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:742e55390d99272bb3191a19443b52d94ed533d46defd05dde24191e9d17452e
|
| 3 |
+
size 623
|
checkpoint-40/trainer_state.json
ADDED
|
@@ -0,0 +1,302 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6572880744934082,
|
| 3 |
+
"best_model_checkpoint": "segformer-b0-finetuned-segments-sidewalk-outputs/checkpoint-40",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 40,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.25,
|
| 12 |
+
"learning_rate": 5.9250000000000004e-05,
|
| 13 |
+
"loss": 1.4918,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.5,
|
| 18 |
+
"learning_rate": 5.85e-05,
|
| 19 |
+
"loss": 1.2963,
|
| 20 |
+
"step": 2
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.75,
|
| 24 |
+
"learning_rate": 5.775e-05,
|
| 25 |
+
"loss": 1.311,
|
| 26 |
+
"step": 3
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.0,
|
| 30 |
+
"learning_rate": 5.6999999999999996e-05,
|
| 31 |
+
"loss": 1.2207,
|
| 32 |
+
"step": 4
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.25,
|
| 36 |
+
"learning_rate": 5.625e-05,
|
| 37 |
+
"loss": 1.2096,
|
| 38 |
+
"step": 5
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.5,
|
| 42 |
+
"learning_rate": 5.550000000000001e-05,
|
| 43 |
+
"loss": 1.1354,
|
| 44 |
+
"step": 6
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.75,
|
| 48 |
+
"learning_rate": 5.475e-05,
|
| 49 |
+
"loss": 1.0495,
|
| 50 |
+
"step": 7
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.0,
|
| 54 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 55 |
+
"loss": 1.1637,
|
| 56 |
+
"step": 8
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 2.25,
|
| 60 |
+
"learning_rate": 5.325e-05,
|
| 61 |
+
"loss": 0.894,
|
| 62 |
+
"step": 9
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.5,
|
| 66 |
+
"learning_rate": 5.25e-05,
|
| 67 |
+
"loss": 0.9463,
|
| 68 |
+
"step": 10
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.75,
|
| 72 |
+
"learning_rate": 5.1750000000000004e-05,
|
| 73 |
+
"loss": 1.1739,
|
| 74 |
+
"step": 11
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 3.0,
|
| 78 |
+
"learning_rate": 5.1e-05,
|
| 79 |
+
"loss": 0.8987,
|
| 80 |
+
"step": 12
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 3.25,
|
| 84 |
+
"learning_rate": 5.025e-05,
|
| 85 |
+
"loss": 0.9905,
|
| 86 |
+
"step": 13
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.5,
|
| 90 |
+
"learning_rate": 4.95e-05,
|
| 91 |
+
"loss": 0.7884,
|
| 92 |
+
"step": 14
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.75,
|
| 96 |
+
"learning_rate": 4.875e-05,
|
| 97 |
+
"loss": 0.7571,
|
| 98 |
+
"step": 15
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 4.0,
|
| 102 |
+
"learning_rate": 4.8e-05,
|
| 103 |
+
"loss": 0.8714,
|
| 104 |
+
"step": 16
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 4.25,
|
| 108 |
+
"learning_rate": 4.7249999999999997e-05,
|
| 109 |
+
"loss": 0.6458,
|
| 110 |
+
"step": 17
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 4.5,
|
| 114 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 115 |
+
"loss": 0.9995,
|
| 116 |
+
"step": 18
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 4.75,
|
| 120 |
+
"learning_rate": 4.575e-05,
|
| 121 |
+
"loss": 0.7726,
|
| 122 |
+
"step": 19
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 5.0,
|
| 126 |
+
"learning_rate": 4.5e-05,
|
| 127 |
+
"loss": 0.8034,
|
| 128 |
+
"step": 20
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 5.0,
|
| 132 |
+
"eval_loss": 1.0316002368927002,
|
| 133 |
+
"eval_mean_accuracy": 0.7020655253953606,
|
| 134 |
+
"eval_mean_iou": 0.364862905521422,
|
| 135 |
+
"eval_overall_accuracy": 0.6520662124792448,
|
| 136 |
+
"eval_per_category_accuracy": [
|
| 137 |
+
NaN,
|
| 138 |
+
0.41182481182481184,
|
| 139 |
+
0.7103931261770244,
|
| 140 |
+
0.9839786381842457
|
| 141 |
+
],
|
| 142 |
+
"eval_per_category_iou": [
|
| 143 |
+
0.0,
|
| 144 |
+
0.41003696685408964,
|
| 145 |
+
0.6273174914575999,
|
| 146 |
+
0.42209716377399836
|
| 147 |
+
],
|
| 148 |
+
"eval_runtime": 2.5163,
|
| 149 |
+
"eval_samples_per_second": 0.795,
|
| 150 |
+
"eval_steps_per_second": 0.397,
|
| 151 |
+
"step": 20
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 5.25,
|
| 155 |
+
"learning_rate": 4.4250000000000005e-05,
|
| 156 |
+
"loss": 0.814,
|
| 157 |
+
"step": 21
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 5.5,
|
| 161 |
+
"learning_rate": 4.35e-05,
|
| 162 |
+
"loss": 0.5981,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 5.75,
|
| 167 |
+
"learning_rate": 4.275e-05,
|
| 168 |
+
"loss": 0.6839,
|
| 169 |
+
"step": 23
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 6.0,
|
| 173 |
+
"learning_rate": 4.2e-05,
|
| 174 |
+
"loss": 0.6713,
|
| 175 |
+
"step": 24
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 6.25,
|
| 179 |
+
"learning_rate": 4.125e-05,
|
| 180 |
+
"loss": 0.7819,
|
| 181 |
+
"step": 25
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 6.5,
|
| 185 |
+
"learning_rate": 4.05e-05,
|
| 186 |
+
"loss": 0.6704,
|
| 187 |
+
"step": 26
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 6.75,
|
| 191 |
+
"learning_rate": 3.975e-05,
|
| 192 |
+
"loss": 0.6488,
|
| 193 |
+
"step": 27
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 7.0,
|
| 197 |
+
"learning_rate": 3.9e-05,
|
| 198 |
+
"loss": 0.6917,
|
| 199 |
+
"step": 28
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 7.25,
|
| 203 |
+
"learning_rate": 3.825e-05,
|
| 204 |
+
"loss": 0.5342,
|
| 205 |
+
"step": 29
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.5,
|
| 209 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 210 |
+
"loss": 0.9604,
|
| 211 |
+
"step": 30
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 7.75,
|
| 215 |
+
"learning_rate": 3.6750000000000006e-05,
|
| 216 |
+
"loss": 0.682,
|
| 217 |
+
"step": 31
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 8.0,
|
| 221 |
+
"learning_rate": 3.6e-05,
|
| 222 |
+
"loss": 0.4576,
|
| 223 |
+
"step": 32
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 8.25,
|
| 227 |
+
"learning_rate": 3.525e-05,
|
| 228 |
+
"loss": 0.5357,
|
| 229 |
+
"step": 33
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 8.5,
|
| 233 |
+
"learning_rate": 3.45e-05,
|
| 234 |
+
"loss": 0.5737,
|
| 235 |
+
"step": 34
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 8.75,
|
| 239 |
+
"learning_rate": 3.375e-05,
|
| 240 |
+
"loss": 0.5618,
|
| 241 |
+
"step": 35
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 9.0,
|
| 245 |
+
"learning_rate": 3.3e-05,
|
| 246 |
+
"loss": 0.5151,
|
| 247 |
+
"step": 36
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 9.25,
|
| 251 |
+
"learning_rate": 3.225e-05,
|
| 252 |
+
"loss": 0.6138,
|
| 253 |
+
"step": 37
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 9.5,
|
| 257 |
+
"learning_rate": 3.15e-05,
|
| 258 |
+
"loss": 0.7417,
|
| 259 |
+
"step": 38
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 9.75,
|
| 263 |
+
"learning_rate": 3.0749999999999995e-05,
|
| 264 |
+
"loss": 0.5539,
|
| 265 |
+
"step": 39
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 10.0,
|
| 269 |
+
"learning_rate": 3e-05,
|
| 270 |
+
"loss": 0.5346,
|
| 271 |
+
"step": 40
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 10.0,
|
| 275 |
+
"eval_loss": 0.6572880744934082,
|
| 276 |
+
"eval_mean_accuracy": 0.8304500516094228,
|
| 277 |
+
"eval_mean_iou": 0.512457610913599,
|
| 278 |
+
"eval_overall_accuracy": 0.8077099766100484,
|
| 279 |
+
"eval_per_category_accuracy": [
|
| 280 |
+
NaN,
|
| 281 |
+
0.7121329121329122,
|
| 282 |
+
0.8221295315442562,
|
| 283 |
+
0.9570877111511
|
| 284 |
+
],
|
| 285 |
+
"eval_per_category_iou": [
|
| 286 |
+
0.0,
|
| 287 |
+
0.7026242881329845,
|
| 288 |
+
0.725572305974316,
|
| 289 |
+
0.6216338495470954
|
| 290 |
+
],
|
| 291 |
+
"eval_runtime": 2.5198,
|
| 292 |
+
"eval_samples_per_second": 0.794,
|
| 293 |
+
"eval_steps_per_second": 0.397,
|
| 294 |
+
"step": 40
|
| 295 |
+
}
|
| 296 |
+
],
|
| 297 |
+
"max_steps": 80,
|
| 298 |
+
"num_train_epochs": 20,
|
| 299 |
+
"total_flos": 1402430470225920.0,
|
| 300 |
+
"trial_name": null,
|
| 301 |
+
"trial_params": null
|
| 302 |
+
}
|
checkpoint-40/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a675845b6e32d3b9ace98a08c50eaa7a24773b2f41b810ad4d204f484f2b0c8
|
| 3 |
+
size 3375
|
checkpoint-60/config.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "nvidia/mit-b0",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SegformerForSemanticSegmentation"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"classifier_dropout_prob": 0.1,
|
| 8 |
+
"decoder_hidden_size": 256,
|
| 9 |
+
"depths": [
|
| 10 |
+
2,
|
| 11 |
+
2,
|
| 12 |
+
2,
|
| 13 |
+
2
|
| 14 |
+
],
|
| 15 |
+
"downsampling_rates": [
|
| 16 |
+
1,
|
| 17 |
+
4,
|
| 18 |
+
8,
|
| 19 |
+
16
|
| 20 |
+
],
|
| 21 |
+
"drop_path_rate": 0.1,
|
| 22 |
+
"hidden_act": "gelu",
|
| 23 |
+
"hidden_dropout_prob": 0.0,
|
| 24 |
+
"hidden_sizes": [
|
| 25 |
+
32,
|
| 26 |
+
64,
|
| 27 |
+
160,
|
| 28 |
+
256
|
| 29 |
+
],
|
| 30 |
+
"id2label": {
|
| 31 |
+
"0": "Nothing",
|
| 32 |
+
"1": "Sand",
|
| 33 |
+
"2": "Soil",
|
| 34 |
+
"3": "Rock"
|
| 35 |
+
},
|
| 36 |
+
"image_size": 224,
|
| 37 |
+
"initializer_range": 0.02,
|
| 38 |
+
"label2id": {
|
| 39 |
+
"Nothing": 0,
|
| 40 |
+
"Rock": 3,
|
| 41 |
+
"Sand": 1,
|
| 42 |
+
"Soil": 2
|
| 43 |
+
},
|
| 44 |
+
"layer_norm_eps": 1e-06,
|
| 45 |
+
"mlp_ratios": [
|
| 46 |
+
4,
|
| 47 |
+
4,
|
| 48 |
+
4,
|
| 49 |
+
4
|
| 50 |
+
],
|
| 51 |
+
"model_type": "segformer",
|
| 52 |
+
"num_attention_heads": [
|
| 53 |
+
1,
|
| 54 |
+
2,
|
| 55 |
+
5,
|
| 56 |
+
8
|
| 57 |
+
],
|
| 58 |
+
"num_channels": 3,
|
| 59 |
+
"num_encoder_blocks": 4,
|
| 60 |
+
"patch_sizes": [
|
| 61 |
+
7,
|
| 62 |
+
3,
|
| 63 |
+
3,
|
| 64 |
+
3
|
| 65 |
+
],
|
| 66 |
+
"reshape_last_stage": true,
|
| 67 |
+
"semantic_loss_ignore_index": 255,
|
| 68 |
+
"sr_ratios": [
|
| 69 |
+
8,
|
| 70 |
+
4,
|
| 71 |
+
2,
|
| 72 |
+
1
|
| 73 |
+
],
|
| 74 |
+
"strides": [
|
| 75 |
+
4,
|
| 76 |
+
2,
|
| 77 |
+
2,
|
| 78 |
+
2
|
| 79 |
+
],
|
| 80 |
+
"torch_dtype": "float32",
|
| 81 |
+
"transformers_version": "4.20.1"
|
| 82 |
+
}
|
checkpoint-60/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2f64ae0a9a15e868036d85d8f0216e77a6dd92527b8192bafffb1abfd254013c
|
| 3 |
+
size 29839501
|
checkpoint-60/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:52b16a8815b3a864fc9d6441699252ee1260a2ae4908fcf0bc3cd23542c709e7
|
| 3 |
+
size 14932001
|
checkpoint-60/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8419031407aedaac7649dfe89911fd6272cafcf9c512e514cc85b16040ac2c6c
|
| 3 |
+
size 14503
|
checkpoint-60/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2f0643f1c92483306dc79d1387c375084f468f477765cb244da631f09cc3ad4
|
| 3 |
+
size 623
|
checkpoint-60/trainer_state.json
ADDED
|
@@ -0,0 +1,445 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5166975259780884,
|
| 3 |
+
"best_model_checkpoint": "segformer-b0-finetuned-segments-sidewalk-outputs/checkpoint-60",
|
| 4 |
+
"epoch": 15.0,
|
| 5 |
+
"global_step": 60,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.25,
|
| 12 |
+
"learning_rate": 5.9250000000000004e-05,
|
| 13 |
+
"loss": 1.4918,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.5,
|
| 18 |
+
"learning_rate": 5.85e-05,
|
| 19 |
+
"loss": 1.2963,
|
| 20 |
+
"step": 2
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.75,
|
| 24 |
+
"learning_rate": 5.775e-05,
|
| 25 |
+
"loss": 1.311,
|
| 26 |
+
"step": 3
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.0,
|
| 30 |
+
"learning_rate": 5.6999999999999996e-05,
|
| 31 |
+
"loss": 1.2207,
|
| 32 |
+
"step": 4
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.25,
|
| 36 |
+
"learning_rate": 5.625e-05,
|
| 37 |
+
"loss": 1.2096,
|
| 38 |
+
"step": 5
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.5,
|
| 42 |
+
"learning_rate": 5.550000000000001e-05,
|
| 43 |
+
"loss": 1.1354,
|
| 44 |
+
"step": 6
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.75,
|
| 48 |
+
"learning_rate": 5.475e-05,
|
| 49 |
+
"loss": 1.0495,
|
| 50 |
+
"step": 7
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.0,
|
| 54 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 55 |
+
"loss": 1.1637,
|
| 56 |
+
"step": 8
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 2.25,
|
| 60 |
+
"learning_rate": 5.325e-05,
|
| 61 |
+
"loss": 0.894,
|
| 62 |
+
"step": 9
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.5,
|
| 66 |
+
"learning_rate": 5.25e-05,
|
| 67 |
+
"loss": 0.9463,
|
| 68 |
+
"step": 10
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.75,
|
| 72 |
+
"learning_rate": 5.1750000000000004e-05,
|
| 73 |
+
"loss": 1.1739,
|
| 74 |
+
"step": 11
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 3.0,
|
| 78 |
+
"learning_rate": 5.1e-05,
|
| 79 |
+
"loss": 0.8987,
|
| 80 |
+
"step": 12
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 3.25,
|
| 84 |
+
"learning_rate": 5.025e-05,
|
| 85 |
+
"loss": 0.9905,
|
| 86 |
+
"step": 13
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.5,
|
| 90 |
+
"learning_rate": 4.95e-05,
|
| 91 |
+
"loss": 0.7884,
|
| 92 |
+
"step": 14
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.75,
|
| 96 |
+
"learning_rate": 4.875e-05,
|
| 97 |
+
"loss": 0.7571,
|
| 98 |
+
"step": 15
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 4.0,
|
| 102 |
+
"learning_rate": 4.8e-05,
|
| 103 |
+
"loss": 0.8714,
|
| 104 |
+
"step": 16
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 4.25,
|
| 108 |
+
"learning_rate": 4.7249999999999997e-05,
|
| 109 |
+
"loss": 0.6458,
|
| 110 |
+
"step": 17
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 4.5,
|
| 114 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 115 |
+
"loss": 0.9995,
|
| 116 |
+
"step": 18
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 4.75,
|
| 120 |
+
"learning_rate": 4.575e-05,
|
| 121 |
+
"loss": 0.7726,
|
| 122 |
+
"step": 19
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 5.0,
|
| 126 |
+
"learning_rate": 4.5e-05,
|
| 127 |
+
"loss": 0.8034,
|
| 128 |
+
"step": 20
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 5.0,
|
| 132 |
+
"eval_loss": 1.0316002368927002,
|
| 133 |
+
"eval_mean_accuracy": 0.7020655253953606,
|
| 134 |
+
"eval_mean_iou": 0.364862905521422,
|
| 135 |
+
"eval_overall_accuracy": 0.6520662124792448,
|
| 136 |
+
"eval_per_category_accuracy": [
|
| 137 |
+
NaN,
|
| 138 |
+
0.41182481182481184,
|
| 139 |
+
0.7103931261770244,
|
| 140 |
+
0.9839786381842457
|
| 141 |
+
],
|
| 142 |
+
"eval_per_category_iou": [
|
| 143 |
+
0.0,
|
| 144 |
+
0.41003696685408964,
|
| 145 |
+
0.6273174914575999,
|
| 146 |
+
0.42209716377399836
|
| 147 |
+
],
|
| 148 |
+
"eval_runtime": 2.5163,
|
| 149 |
+
"eval_samples_per_second": 0.795,
|
| 150 |
+
"eval_steps_per_second": 0.397,
|
| 151 |
+
"step": 20
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 5.25,
|
| 155 |
+
"learning_rate": 4.4250000000000005e-05,
|
| 156 |
+
"loss": 0.814,
|
| 157 |
+
"step": 21
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 5.5,
|
| 161 |
+
"learning_rate": 4.35e-05,
|
| 162 |
+
"loss": 0.5981,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 5.75,
|
| 167 |
+
"learning_rate": 4.275e-05,
|
| 168 |
+
"loss": 0.6839,
|
| 169 |
+
"step": 23
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 6.0,
|
| 173 |
+
"learning_rate": 4.2e-05,
|
| 174 |
+
"loss": 0.6713,
|
| 175 |
+
"step": 24
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 6.25,
|
| 179 |
+
"learning_rate": 4.125e-05,
|
| 180 |
+
"loss": 0.7819,
|
| 181 |
+
"step": 25
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 6.5,
|
| 185 |
+
"learning_rate": 4.05e-05,
|
| 186 |
+
"loss": 0.6704,
|
| 187 |
+
"step": 26
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 6.75,
|
| 191 |
+
"learning_rate": 3.975e-05,
|
| 192 |
+
"loss": 0.6488,
|
| 193 |
+
"step": 27
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 7.0,
|
| 197 |
+
"learning_rate": 3.9e-05,
|
| 198 |
+
"loss": 0.6917,
|
| 199 |
+
"step": 28
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 7.25,
|
| 203 |
+
"learning_rate": 3.825e-05,
|
| 204 |
+
"loss": 0.5342,
|
| 205 |
+
"step": 29
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.5,
|
| 209 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 210 |
+
"loss": 0.9604,
|
| 211 |
+
"step": 30
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 7.75,
|
| 215 |
+
"learning_rate": 3.6750000000000006e-05,
|
| 216 |
+
"loss": 0.682,
|
| 217 |
+
"step": 31
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 8.0,
|
| 221 |
+
"learning_rate": 3.6e-05,
|
| 222 |
+
"loss": 0.4576,
|
| 223 |
+
"step": 32
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 8.25,
|
| 227 |
+
"learning_rate": 3.525e-05,
|
| 228 |
+
"loss": 0.5357,
|
| 229 |
+
"step": 33
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 8.5,
|
| 233 |
+
"learning_rate": 3.45e-05,
|
| 234 |
+
"loss": 0.5737,
|
| 235 |
+
"step": 34
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 8.75,
|
| 239 |
+
"learning_rate": 3.375e-05,
|
| 240 |
+
"loss": 0.5618,
|
| 241 |
+
"step": 35
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 9.0,
|
| 245 |
+
"learning_rate": 3.3e-05,
|
| 246 |
+
"loss": 0.5151,
|
| 247 |
+
"step": 36
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 9.25,
|
| 251 |
+
"learning_rate": 3.225e-05,
|
| 252 |
+
"loss": 0.6138,
|
| 253 |
+
"step": 37
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 9.5,
|
| 257 |
+
"learning_rate": 3.15e-05,
|
| 258 |
+
"loss": 0.7417,
|
| 259 |
+
"step": 38
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 9.75,
|
| 263 |
+
"learning_rate": 3.0749999999999995e-05,
|
| 264 |
+
"loss": 0.5539,
|
| 265 |
+
"step": 39
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 10.0,
|
| 269 |
+
"learning_rate": 3e-05,
|
| 270 |
+
"loss": 0.5346,
|
| 271 |
+
"step": 40
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 10.0,
|
| 275 |
+
"eval_loss": 0.6572880744934082,
|
| 276 |
+
"eval_mean_accuracy": 0.8304500516094228,
|
| 277 |
+
"eval_mean_iou": 0.512457610913599,
|
| 278 |
+
"eval_overall_accuracy": 0.8077099766100484,
|
| 279 |
+
"eval_per_category_accuracy": [
|
| 280 |
+
NaN,
|
| 281 |
+
0.7121329121329122,
|
| 282 |
+
0.8221295315442562,
|
| 283 |
+
0.9570877111511
|
| 284 |
+
],
|
| 285 |
+
"eval_per_category_iou": [
|
| 286 |
+
0.0,
|
| 287 |
+
0.7026242881329845,
|
| 288 |
+
0.725572305974316,
|
| 289 |
+
0.6216338495470954
|
| 290 |
+
],
|
| 291 |
+
"eval_runtime": 2.5198,
|
| 292 |
+
"eval_samples_per_second": 0.794,
|
| 293 |
+
"eval_steps_per_second": 0.397,
|
| 294 |
+
"step": 40
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 10.25,
|
| 298 |
+
"learning_rate": 2.925e-05,
|
| 299 |
+
"loss": 0.5535,
|
| 300 |
+
"step": 41
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 10.5,
|
| 304 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 305 |
+
"loss": 0.6405,
|
| 306 |
+
"step": 42
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 10.75,
|
| 310 |
+
"learning_rate": 2.7750000000000004e-05,
|
| 311 |
+
"loss": 0.594,
|
| 312 |
+
"step": 43
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 11.0,
|
| 316 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 317 |
+
"loss": 0.7267,
|
| 318 |
+
"step": 44
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 11.25,
|
| 322 |
+
"learning_rate": 2.625e-05,
|
| 323 |
+
"loss": 0.6504,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 11.5,
|
| 328 |
+
"learning_rate": 2.55e-05,
|
| 329 |
+
"loss": 0.7008,
|
| 330 |
+
"step": 46
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 11.75,
|
| 334 |
+
"learning_rate": 2.475e-05,
|
| 335 |
+
"loss": 0.6233,
|
| 336 |
+
"step": 47
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 12.0,
|
| 340 |
+
"learning_rate": 2.4e-05,
|
| 341 |
+
"loss": 0.474,
|
| 342 |
+
"step": 48
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 12.25,
|
| 346 |
+
"learning_rate": 2.3250000000000003e-05,
|
| 347 |
+
"loss": 0.5064,
|
| 348 |
+
"step": 49
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 12.5,
|
| 352 |
+
"learning_rate": 2.25e-05,
|
| 353 |
+
"loss": 0.604,
|
| 354 |
+
"step": 50
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 12.75,
|
| 358 |
+
"learning_rate": 2.175e-05,
|
| 359 |
+
"loss": 0.4316,
|
| 360 |
+
"step": 51
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 13.0,
|
| 364 |
+
"learning_rate": 2.1e-05,
|
| 365 |
+
"loss": 0.4736,
|
| 366 |
+
"step": 52
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 13.25,
|
| 370 |
+
"learning_rate": 2.025e-05,
|
| 371 |
+
"loss": 0.5676,
|
| 372 |
+
"step": 53
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 13.5,
|
| 376 |
+
"learning_rate": 1.95e-05,
|
| 377 |
+
"loss": 0.4962,
|
| 378 |
+
"step": 54
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 13.75,
|
| 382 |
+
"learning_rate": 1.8750000000000002e-05,
|
| 383 |
+
"loss": 0.4723,
|
| 384 |
+
"step": 55
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 14.0,
|
| 388 |
+
"learning_rate": 1.8e-05,
|
| 389 |
+
"loss": 0.5382,
|
| 390 |
+
"step": 56
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 14.25,
|
| 394 |
+
"learning_rate": 1.725e-05,
|
| 395 |
+
"loss": 0.3868,
|
| 396 |
+
"step": 57
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 14.5,
|
| 400 |
+
"learning_rate": 1.65e-05,
|
| 401 |
+
"loss": 0.4054,
|
| 402 |
+
"step": 58
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 14.75,
|
| 406 |
+
"learning_rate": 1.575e-05,
|
| 407 |
+
"loss": 0.4154,
|
| 408 |
+
"step": 59
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 15.0,
|
| 412 |
+
"learning_rate": 1.5e-05,
|
| 413 |
+
"loss": 0.4055,
|
| 414 |
+
"step": 60
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 15.0,
|
| 418 |
+
"eval_loss": 0.5166975259780884,
|
| 419 |
+
"eval_mean_accuracy": 0.8371801359160734,
|
| 420 |
+
"eval_mean_iou": 0.514730970731156,
|
| 421 |
+
"eval_overall_accuracy": 0.8129674612444415,
|
| 422 |
+
"eval_per_category_accuracy": [
|
| 423 |
+
NaN,
|
| 424 |
+
0.7417573417573418,
|
| 425 |
+
0.8012888418079096,
|
| 426 |
+
0.9684942241829686
|
| 427 |
+
],
|
| 428 |
+
"eval_per_category_iou": [
|
| 429 |
+
0.0,
|
| 430 |
+
0.73762061369489,
|
| 431 |
+
0.7308618780823296,
|
| 432 |
+
0.5904413911474047
|
| 433 |
+
],
|
| 434 |
+
"eval_runtime": 2.8676,
|
| 435 |
+
"eval_samples_per_second": 0.697,
|
| 436 |
+
"eval_steps_per_second": 0.349,
|
| 437 |
+
"step": 60
|
| 438 |
+
}
|
| 439 |
+
],
|
| 440 |
+
"max_steps": 80,
|
| 441 |
+
"num_train_epochs": 20,
|
| 442 |
+
"total_flos": 2103645705338880.0,
|
| 443 |
+
"trial_name": null,
|
| 444 |
+
"trial_params": null
|
| 445 |
+
}
|
checkpoint-60/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a675845b6e32d3b9ace98a08c50eaa7a24773b2f41b810ad4d204f484f2b0c8
|
| 3 |
+
size 3375
|
checkpoint-80/config.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "nvidia/mit-b0",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SegformerForSemanticSegmentation"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"classifier_dropout_prob": 0.1,
|
| 8 |
+
"decoder_hidden_size": 256,
|
| 9 |
+
"depths": [
|
| 10 |
+
2,
|
| 11 |
+
2,
|
| 12 |
+
2,
|
| 13 |
+
2
|
| 14 |
+
],
|
| 15 |
+
"downsampling_rates": [
|
| 16 |
+
1,
|
| 17 |
+
4,
|
| 18 |
+
8,
|
| 19 |
+
16
|
| 20 |
+
],
|
| 21 |
+
"drop_path_rate": 0.1,
|
| 22 |
+
"hidden_act": "gelu",
|
| 23 |
+
"hidden_dropout_prob": 0.0,
|
| 24 |
+
"hidden_sizes": [
|
| 25 |
+
32,
|
| 26 |
+
64,
|
| 27 |
+
160,
|
| 28 |
+
256
|
| 29 |
+
],
|
| 30 |
+
"id2label": {
|
| 31 |
+
"0": "Nothing",
|
| 32 |
+
"1": "Sand",
|
| 33 |
+
"2": "Soil",
|
| 34 |
+
"3": "Rock"
|
| 35 |
+
},
|
| 36 |
+
"image_size": 224,
|
| 37 |
+
"initializer_range": 0.02,
|
| 38 |
+
"label2id": {
|
| 39 |
+
"Nothing": 0,
|
| 40 |
+
"Rock": 3,
|
| 41 |
+
"Sand": 1,
|
| 42 |
+
"Soil": 2
|
| 43 |
+
},
|
| 44 |
+
"layer_norm_eps": 1e-06,
|
| 45 |
+
"mlp_ratios": [
|
| 46 |
+
4,
|
| 47 |
+
4,
|
| 48 |
+
4,
|
| 49 |
+
4
|
| 50 |
+
],
|
| 51 |
+
"model_type": "segformer",
|
| 52 |
+
"num_attention_heads": [
|
| 53 |
+
1,
|
| 54 |
+
2,
|
| 55 |
+
5,
|
| 56 |
+
8
|
| 57 |
+
],
|
| 58 |
+
"num_channels": 3,
|
| 59 |
+
"num_encoder_blocks": 4,
|
| 60 |
+
"patch_sizes": [
|
| 61 |
+
7,
|
| 62 |
+
3,
|
| 63 |
+
3,
|
| 64 |
+
3
|
| 65 |
+
],
|
| 66 |
+
"reshape_last_stage": true,
|
| 67 |
+
"semantic_loss_ignore_index": 255,
|
| 68 |
+
"sr_ratios": [
|
| 69 |
+
8,
|
| 70 |
+
4,
|
| 71 |
+
2,
|
| 72 |
+
1
|
| 73 |
+
],
|
| 74 |
+
"strides": [
|
| 75 |
+
4,
|
| 76 |
+
2,
|
| 77 |
+
2,
|
| 78 |
+
2
|
| 79 |
+
],
|
| 80 |
+
"torch_dtype": "float32",
|
| 81 |
+
"transformers_version": "4.20.1"
|
| 82 |
+
}
|
checkpoint-80/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3320cf3efbbeb7194b83e7573212ea1229a309b4e66d6bcc34a8f743071bc24c
|
| 3 |
+
size 29839501
|
checkpoint-80/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:31f7ec2e536a470667d58a31364ff3c3271eedb4be97967aa2026f9f3c5619df
|
| 3 |
+
size 14932001
|
checkpoint-80/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9acd1dcdfba8e6bf64a39c9f3329e5a2f91e3d793e737992936d4e5521737f5a
|
| 3 |
+
size 14503
|
checkpoint-80/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2e9ccfe2d18f98be6ea663cbc1e44ddd85e93156187cce6b1917005e15eb7f12
|
| 3 |
+
size 623
|
checkpoint-80/trainer_state.json
ADDED
|
@@ -0,0 +1,588 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.4903029203414917,
|
| 3 |
+
"best_model_checkpoint": "segformer-b0-finetuned-segments-sidewalk-outputs/checkpoint-80",
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"global_step": 80,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.25,
|
| 12 |
+
"learning_rate": 5.9250000000000004e-05,
|
| 13 |
+
"loss": 1.4918,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.5,
|
| 18 |
+
"learning_rate": 5.85e-05,
|
| 19 |
+
"loss": 1.2963,
|
| 20 |
+
"step": 2
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.75,
|
| 24 |
+
"learning_rate": 5.775e-05,
|
| 25 |
+
"loss": 1.311,
|
| 26 |
+
"step": 3
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.0,
|
| 30 |
+
"learning_rate": 5.6999999999999996e-05,
|
| 31 |
+
"loss": 1.2207,
|
| 32 |
+
"step": 4
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.25,
|
| 36 |
+
"learning_rate": 5.625e-05,
|
| 37 |
+
"loss": 1.2096,
|
| 38 |
+
"step": 5
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.5,
|
| 42 |
+
"learning_rate": 5.550000000000001e-05,
|
| 43 |
+
"loss": 1.1354,
|
| 44 |
+
"step": 6
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.75,
|
| 48 |
+
"learning_rate": 5.475e-05,
|
| 49 |
+
"loss": 1.0495,
|
| 50 |
+
"step": 7
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.0,
|
| 54 |
+
"learning_rate": 5.4000000000000005e-05,
|
| 55 |
+
"loss": 1.1637,
|
| 56 |
+
"step": 8
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 2.25,
|
| 60 |
+
"learning_rate": 5.325e-05,
|
| 61 |
+
"loss": 0.894,
|
| 62 |
+
"step": 9
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.5,
|
| 66 |
+
"learning_rate": 5.25e-05,
|
| 67 |
+
"loss": 0.9463,
|
| 68 |
+
"step": 10
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.75,
|
| 72 |
+
"learning_rate": 5.1750000000000004e-05,
|
| 73 |
+
"loss": 1.1739,
|
| 74 |
+
"step": 11
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 3.0,
|
| 78 |
+
"learning_rate": 5.1e-05,
|
| 79 |
+
"loss": 0.8987,
|
| 80 |
+
"step": 12
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 3.25,
|
| 84 |
+
"learning_rate": 5.025e-05,
|
| 85 |
+
"loss": 0.9905,
|
| 86 |
+
"step": 13
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.5,
|
| 90 |
+
"learning_rate": 4.95e-05,
|
| 91 |
+
"loss": 0.7884,
|
| 92 |
+
"step": 14
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.75,
|
| 96 |
+
"learning_rate": 4.875e-05,
|
| 97 |
+
"loss": 0.7571,
|
| 98 |
+
"step": 15
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 4.0,
|
| 102 |
+
"learning_rate": 4.8e-05,
|
| 103 |
+
"loss": 0.8714,
|
| 104 |
+
"step": 16
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 4.25,
|
| 108 |
+
"learning_rate": 4.7249999999999997e-05,
|
| 109 |
+
"loss": 0.6458,
|
| 110 |
+
"step": 17
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 4.5,
|
| 114 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 115 |
+
"loss": 0.9995,
|
| 116 |
+
"step": 18
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 4.75,
|
| 120 |
+
"learning_rate": 4.575e-05,
|
| 121 |
+
"loss": 0.7726,
|
| 122 |
+
"step": 19
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 5.0,
|
| 126 |
+
"learning_rate": 4.5e-05,
|
| 127 |
+
"loss": 0.8034,
|
| 128 |
+
"step": 20
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 5.0,
|
| 132 |
+
"eval_loss": 1.0316002368927002,
|
| 133 |
+
"eval_mean_accuracy": 0.7020655253953606,
|
| 134 |
+
"eval_mean_iou": 0.364862905521422,
|
| 135 |
+
"eval_overall_accuracy": 0.6520662124792448,
|
| 136 |
+
"eval_per_category_accuracy": [
|
| 137 |
+
NaN,
|
| 138 |
+
0.41182481182481184,
|
| 139 |
+
0.7103931261770244,
|
| 140 |
+
0.9839786381842457
|
| 141 |
+
],
|
| 142 |
+
"eval_per_category_iou": [
|
| 143 |
+
0.0,
|
| 144 |
+
0.41003696685408964,
|
| 145 |
+
0.6273174914575999,
|
| 146 |
+
0.42209716377399836
|
| 147 |
+
],
|
| 148 |
+
"eval_runtime": 2.5163,
|
| 149 |
+
"eval_samples_per_second": 0.795,
|
| 150 |
+
"eval_steps_per_second": 0.397,
|
| 151 |
+
"step": 20
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 5.25,
|
| 155 |
+
"learning_rate": 4.4250000000000005e-05,
|
| 156 |
+
"loss": 0.814,
|
| 157 |
+
"step": 21
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 5.5,
|
| 161 |
+
"learning_rate": 4.35e-05,
|
| 162 |
+
"loss": 0.5981,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 5.75,
|
| 167 |
+
"learning_rate": 4.275e-05,
|
| 168 |
+
"loss": 0.6839,
|
| 169 |
+
"step": 23
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 6.0,
|
| 173 |
+
"learning_rate": 4.2e-05,
|
| 174 |
+
"loss": 0.6713,
|
| 175 |
+
"step": 24
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 6.25,
|
| 179 |
+
"learning_rate": 4.125e-05,
|
| 180 |
+
"loss": 0.7819,
|
| 181 |
+
"step": 25
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 6.5,
|
| 185 |
+
"learning_rate": 4.05e-05,
|
| 186 |
+
"loss": 0.6704,
|
| 187 |
+
"step": 26
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 6.75,
|
| 191 |
+
"learning_rate": 3.975e-05,
|
| 192 |
+
"loss": 0.6488,
|
| 193 |
+
"step": 27
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 7.0,
|
| 197 |
+
"learning_rate": 3.9e-05,
|
| 198 |
+
"loss": 0.6917,
|
| 199 |
+
"step": 28
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 7.25,
|
| 203 |
+
"learning_rate": 3.825e-05,
|
| 204 |
+
"loss": 0.5342,
|
| 205 |
+
"step": 29
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.5,
|
| 209 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 210 |
+
"loss": 0.9604,
|
| 211 |
+
"step": 30
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 7.75,
|
| 215 |
+
"learning_rate": 3.6750000000000006e-05,
|
| 216 |
+
"loss": 0.682,
|
| 217 |
+
"step": 31
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 8.0,
|
| 221 |
+
"learning_rate": 3.6e-05,
|
| 222 |
+
"loss": 0.4576,
|
| 223 |
+
"step": 32
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 8.25,
|
| 227 |
+
"learning_rate": 3.525e-05,
|
| 228 |
+
"loss": 0.5357,
|
| 229 |
+
"step": 33
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 8.5,
|
| 233 |
+
"learning_rate": 3.45e-05,
|
| 234 |
+
"loss": 0.5737,
|
| 235 |
+
"step": 34
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 8.75,
|
| 239 |
+
"learning_rate": 3.375e-05,
|
| 240 |
+
"loss": 0.5618,
|
| 241 |
+
"step": 35
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 9.0,
|
| 245 |
+
"learning_rate": 3.3e-05,
|
| 246 |
+
"loss": 0.5151,
|
| 247 |
+
"step": 36
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 9.25,
|
| 251 |
+
"learning_rate": 3.225e-05,
|
| 252 |
+
"loss": 0.6138,
|
| 253 |
+
"step": 37
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 9.5,
|
| 257 |
+
"learning_rate": 3.15e-05,
|
| 258 |
+
"loss": 0.7417,
|
| 259 |
+
"step": 38
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 9.75,
|
| 263 |
+
"learning_rate": 3.0749999999999995e-05,
|
| 264 |
+
"loss": 0.5539,
|
| 265 |
+
"step": 39
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 10.0,
|
| 269 |
+
"learning_rate": 3e-05,
|
| 270 |
+
"loss": 0.5346,
|
| 271 |
+
"step": 40
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 10.0,
|
| 275 |
+
"eval_loss": 0.6572880744934082,
|
| 276 |
+
"eval_mean_accuracy": 0.8304500516094228,
|
| 277 |
+
"eval_mean_iou": 0.512457610913599,
|
| 278 |
+
"eval_overall_accuracy": 0.8077099766100484,
|
| 279 |
+
"eval_per_category_accuracy": [
|
| 280 |
+
NaN,
|
| 281 |
+
0.7121329121329122,
|
| 282 |
+
0.8221295315442562,
|
| 283 |
+
0.9570877111511
|
| 284 |
+
],
|
| 285 |
+
"eval_per_category_iou": [
|
| 286 |
+
0.0,
|
| 287 |
+
0.7026242881329845,
|
| 288 |
+
0.725572305974316,
|
| 289 |
+
0.6216338495470954
|
| 290 |
+
],
|
| 291 |
+
"eval_runtime": 2.5198,
|
| 292 |
+
"eval_samples_per_second": 0.794,
|
| 293 |
+
"eval_steps_per_second": 0.397,
|
| 294 |
+
"step": 40
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 10.25,
|
| 298 |
+
"learning_rate": 2.925e-05,
|
| 299 |
+
"loss": 0.5535,
|
| 300 |
+
"step": 41
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 10.5,
|
| 304 |
+
"learning_rate": 2.8499999999999998e-05,
|
| 305 |
+
"loss": 0.6405,
|
| 306 |
+
"step": 42
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 10.75,
|
| 310 |
+
"learning_rate": 2.7750000000000004e-05,
|
| 311 |
+
"loss": 0.594,
|
| 312 |
+
"step": 43
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 11.0,
|
| 316 |
+
"learning_rate": 2.7000000000000002e-05,
|
| 317 |
+
"loss": 0.7267,
|
| 318 |
+
"step": 44
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 11.25,
|
| 322 |
+
"learning_rate": 2.625e-05,
|
| 323 |
+
"loss": 0.6504,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 11.5,
|
| 328 |
+
"learning_rate": 2.55e-05,
|
| 329 |
+
"loss": 0.7008,
|
| 330 |
+
"step": 46
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 11.75,
|
| 334 |
+
"learning_rate": 2.475e-05,
|
| 335 |
+
"loss": 0.6233,
|
| 336 |
+
"step": 47
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 12.0,
|
| 340 |
+
"learning_rate": 2.4e-05,
|
| 341 |
+
"loss": 0.474,
|
| 342 |
+
"step": 48
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 12.25,
|
| 346 |
+
"learning_rate": 2.3250000000000003e-05,
|
| 347 |
+
"loss": 0.5064,
|
| 348 |
+
"step": 49
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 12.5,
|
| 352 |
+
"learning_rate": 2.25e-05,
|
| 353 |
+
"loss": 0.604,
|
| 354 |
+
"step": 50
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 12.75,
|
| 358 |
+
"learning_rate": 2.175e-05,
|
| 359 |
+
"loss": 0.4316,
|
| 360 |
+
"step": 51
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 13.0,
|
| 364 |
+
"learning_rate": 2.1e-05,
|
| 365 |
+
"loss": 0.4736,
|
| 366 |
+
"step": 52
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 13.25,
|
| 370 |
+
"learning_rate": 2.025e-05,
|
| 371 |
+
"loss": 0.5676,
|
| 372 |
+
"step": 53
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 13.5,
|
| 376 |
+
"learning_rate": 1.95e-05,
|
| 377 |
+
"loss": 0.4962,
|
| 378 |
+
"step": 54
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 13.75,
|
| 382 |
+
"learning_rate": 1.8750000000000002e-05,
|
| 383 |
+
"loss": 0.4723,
|
| 384 |
+
"step": 55
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 14.0,
|
| 388 |
+
"learning_rate": 1.8e-05,
|
| 389 |
+
"loss": 0.5382,
|
| 390 |
+
"step": 56
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 14.25,
|
| 394 |
+
"learning_rate": 1.725e-05,
|
| 395 |
+
"loss": 0.3868,
|
| 396 |
+
"step": 57
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 14.5,
|
| 400 |
+
"learning_rate": 1.65e-05,
|
| 401 |
+
"loss": 0.4054,
|
| 402 |
+
"step": 58
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 14.75,
|
| 406 |
+
"learning_rate": 1.575e-05,
|
| 407 |
+
"loss": 0.4154,
|
| 408 |
+
"step": 59
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 15.0,
|
| 412 |
+
"learning_rate": 1.5e-05,
|
| 413 |
+
"loss": 0.4055,
|
| 414 |
+
"step": 60
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 15.0,
|
| 418 |
+
"eval_loss": 0.5166975259780884,
|
| 419 |
+
"eval_mean_accuracy": 0.8371801359160734,
|
| 420 |
+
"eval_mean_iou": 0.514730970731156,
|
| 421 |
+
"eval_overall_accuracy": 0.8129674612444415,
|
| 422 |
+
"eval_per_category_accuracy": [
|
| 423 |
+
NaN,
|
| 424 |
+
0.7417573417573418,
|
| 425 |
+
0.8012888418079096,
|
| 426 |
+
0.9684942241829686
|
| 427 |
+
],
|
| 428 |
+
"eval_per_category_iou": [
|
| 429 |
+
0.0,
|
| 430 |
+
0.73762061369489,
|
| 431 |
+
0.7308618780823296,
|
| 432 |
+
0.5904413911474047
|
| 433 |
+
],
|
| 434 |
+
"eval_runtime": 2.8676,
|
| 435 |
+
"eval_samples_per_second": 0.697,
|
| 436 |
+
"eval_steps_per_second": 0.349,
|
| 437 |
+
"step": 60
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 15.25,
|
| 441 |
+
"learning_rate": 1.4249999999999999e-05,
|
| 442 |
+
"loss": 0.7572,
|
| 443 |
+
"step": 61
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 15.5,
|
| 447 |
+
"learning_rate": 1.3500000000000001e-05,
|
| 448 |
+
"loss": 0.415,
|
| 449 |
+
"step": 62
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 15.75,
|
| 453 |
+
"learning_rate": 1.275e-05,
|
| 454 |
+
"loss": 0.4676,
|
| 455 |
+
"step": 63
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 16.0,
|
| 459 |
+
"learning_rate": 1.2e-05,
|
| 460 |
+
"loss": 0.4485,
|
| 461 |
+
"step": 64
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 16.25,
|
| 465 |
+
"learning_rate": 1.125e-05,
|
| 466 |
+
"loss": 0.4861,
|
| 467 |
+
"step": 65
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 16.5,
|
| 471 |
+
"learning_rate": 1.05e-05,
|
| 472 |
+
"loss": 0.4819,
|
| 473 |
+
"step": 66
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 16.75,
|
| 477 |
+
"learning_rate": 9.75e-06,
|
| 478 |
+
"loss": 0.3281,
|
| 479 |
+
"step": 67
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 17.0,
|
| 483 |
+
"learning_rate": 9e-06,
|
| 484 |
+
"loss": 0.4026,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 17.25,
|
| 489 |
+
"learning_rate": 8.25e-06,
|
| 490 |
+
"loss": 0.3895,
|
| 491 |
+
"step": 69
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 17.5,
|
| 495 |
+
"learning_rate": 7.5e-06,
|
| 496 |
+
"loss": 0.5976,
|
| 497 |
+
"step": 70
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 17.75,
|
| 501 |
+
"learning_rate": 6.750000000000001e-06,
|
| 502 |
+
"loss": 0.6719,
|
| 503 |
+
"step": 71
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 18.0,
|
| 507 |
+
"learning_rate": 6e-06,
|
| 508 |
+
"loss": 0.5351,
|
| 509 |
+
"step": 72
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 18.25,
|
| 513 |
+
"learning_rate": 5.25e-06,
|
| 514 |
+
"loss": 0.5423,
|
| 515 |
+
"step": 73
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 18.5,
|
| 519 |
+
"learning_rate": 4.5e-06,
|
| 520 |
+
"loss": 0.342,
|
| 521 |
+
"step": 74
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 18.75,
|
| 525 |
+
"learning_rate": 3.75e-06,
|
| 526 |
+
"loss": 0.4317,
|
| 527 |
+
"step": 75
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 19.0,
|
| 531 |
+
"learning_rate": 3e-06,
|
| 532 |
+
"loss": 0.4026,
|
| 533 |
+
"step": 76
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 19.25,
|
| 537 |
+
"learning_rate": 2.25e-06,
|
| 538 |
+
"loss": 0.3905,
|
| 539 |
+
"step": 77
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 19.5,
|
| 543 |
+
"learning_rate": 1.5e-06,
|
| 544 |
+
"loss": 0.808,
|
| 545 |
+
"step": 78
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 19.75,
|
| 549 |
+
"learning_rate": 7.5e-07,
|
| 550 |
+
"loss": 0.4165,
|
| 551 |
+
"step": 79
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 20.0,
|
| 555 |
+
"learning_rate": 0.0,
|
| 556 |
+
"loss": 0.6116,
|
| 557 |
+
"step": 80
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 20.0,
|
| 561 |
+
"eval_loss": 0.4903029203414917,
|
| 562 |
+
"eval_mean_accuracy": 0.8806864679806351,
|
| 563 |
+
"eval_mean_iou": 0.5690081414204357,
|
| 564 |
+
"eval_overall_accuracy": 0.8644673906842537,
|
| 565 |
+
"eval_per_category_accuracy": [
|
| 566 |
+
NaN,
|
| 567 |
+
0.797971997971998,
|
| 568 |
+
0.873271245291902,
|
| 569 |
+
0.9708161606780055
|
| 570 |
+
],
|
| 571 |
+
"eval_per_category_iou": [
|
| 572 |
+
0.0,
|
| 573 |
+
0.7928883101986406,
|
| 574 |
+
0.8130753424657534,
|
| 575 |
+
0.6700689130173485
|
| 576 |
+
],
|
| 577 |
+
"eval_runtime": 2.4992,
|
| 578 |
+
"eval_samples_per_second": 0.8,
|
| 579 |
+
"eval_steps_per_second": 0.4,
|
| 580 |
+
"step": 80
|
| 581 |
+
}
|
| 582 |
+
],
|
| 583 |
+
"max_steps": 80,
|
| 584 |
+
"num_train_epochs": 20,
|
| 585 |
+
"total_flos": 2804860940451840.0,
|
| 586 |
+
"trial_name": null,
|
| 587 |
+
"trial_params": null
|
| 588 |
+
}
|
checkpoint-80/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a675845b6e32d3b9ace98a08c50eaa7a24773b2f41b810ad4d204f484f2b0c8
|
| 3 |
+
size 3375
|
config.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "nvidia/mit-b0",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SegformerForSemanticSegmentation"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"classifier_dropout_prob": 0.1,
|
| 8 |
+
"decoder_hidden_size": 256,
|
| 9 |
+
"depths": [
|
| 10 |
+
2,
|
| 11 |
+
2,
|
| 12 |
+
2,
|
| 13 |
+
2
|
| 14 |
+
],
|
| 15 |
+
"downsampling_rates": [
|
| 16 |
+
1,
|
| 17 |
+
4,
|
| 18 |
+
8,
|
| 19 |
+
16
|
| 20 |
+
],
|
| 21 |
+
"drop_path_rate": 0.1,
|
| 22 |
+
"hidden_act": "gelu",
|
| 23 |
+
"hidden_dropout_prob": 0.0,
|
| 24 |
+
"hidden_sizes": [
|
| 25 |
+
32,
|
| 26 |
+
64,
|
| 27 |
+
160,
|
| 28 |
+
256
|
| 29 |
+
],
|
| 30 |
+
"id2label": {
|
| 31 |
+
"0": "Nothing",
|
| 32 |
+
"1": "Sand",
|
| 33 |
+
"2": "Soil",
|
| 34 |
+
"3": "Rock"
|
| 35 |
+
},
|
| 36 |
+
"image_size": 224,
|
| 37 |
+
"initializer_range": 0.02,
|
| 38 |
+
"label2id": {
|
| 39 |
+
"Nothing": 0,
|
| 40 |
+
"Rock": 3,
|
| 41 |
+
"Sand": 1,
|
| 42 |
+
"Soil": 2
|
| 43 |
+
},
|
| 44 |
+
"layer_norm_eps": 1e-06,
|
| 45 |
+
"mlp_ratios": [
|
| 46 |
+
4,
|
| 47 |
+
4,
|
| 48 |
+
4,
|
| 49 |
+
4
|
| 50 |
+
],
|
| 51 |
+
"model_type": "segformer",
|
| 52 |
+
"num_attention_heads": [
|
| 53 |
+
1,
|
| 54 |
+
2,
|
| 55 |
+
5,
|
| 56 |
+
8
|
| 57 |
+
],
|
| 58 |
+
"num_channels": 3,
|
| 59 |
+
"num_encoder_blocks": 4,
|
| 60 |
+
"patch_sizes": [
|
| 61 |
+
7,
|
| 62 |
+
3,
|
| 63 |
+
3,
|
| 64 |
+
3
|
| 65 |
+
],
|
| 66 |
+
"reshape_last_stage": true,
|
| 67 |
+
"semantic_loss_ignore_index": 255,
|
| 68 |
+
"sr_ratios": [
|
| 69 |
+
8,
|
| 70 |
+
4,
|
| 71 |
+
2,
|
| 72 |
+
1
|
| 73 |
+
],
|
| 74 |
+
"strides": [
|
| 75 |
+
4,
|
| 76 |
+
2,
|
| 77 |
+
2,
|
| 78 |
+
2
|
| 79 |
+
],
|
| 80 |
+
"torch_dtype": "float32",
|
| 81 |
+
"transformers_version": "4.20.1"
|
| 82 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:31f7ec2e536a470667d58a31364ff3c3271eedb4be97967aa2026f9f3c5619df
|
| 3 |
+
size 14932001
|
runs/Jul14_06-17-20_a2366e527b16/1657779466.1818385/events.out.tfevents.1657779466.a2366e527b16.71.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e68c7ab8fdb85f57c4cd562aa865208c7ac27a52f43264559612f7eb28241696
|
| 3 |
+
size 5456
|
runs/Jul14_06-17-20_a2366e527b16/events.out.tfevents.1657779466.a2366e527b16.71.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ee49e4f4e63aa162b8b3f4344cb8f2d8069e5fe307165a181853117043b04df3
|
| 3 |
+
size 18637
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a675845b6e32d3b9ace98a08c50eaa7a24773b2f41b810ad4d204f484f2b0c8
|
| 3 |
+
size 3375
|