Training in progress, step 80
Browse files- checkpoint-80/config.json +80 -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 +564 -0
- checkpoint-80/training_args.bin +3 -0
- config.json +80 -0
- pytorch_model.bin +3 -0
- runs/Feb08_09-49-29_robovision.csres.utexas.edu/events.out.tfevents.1707407376.robovision.csres.utexas.edu.889113.0 +3 -0
- training_args.bin +3 -0
checkpoint-80/config.json
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "nvidia/mit-b5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SegformerForSemanticSegmentation"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"classifier_dropout_prob": 0.1,
|
| 8 |
+
"decoder_hidden_size": 768,
|
| 9 |
+
"depths": [
|
| 10 |
+
3,
|
| 11 |
+
6,
|
| 12 |
+
40,
|
| 13 |
+
3
|
| 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 |
+
64,
|
| 26 |
+
128,
|
| 27 |
+
320,
|
| 28 |
+
512
|
| 29 |
+
],
|
| 30 |
+
"id2label": {
|
| 31 |
+
"0": "unlabeled",
|
| 32 |
+
"1": "safe",
|
| 33 |
+
"2": "unsafe"
|
| 34 |
+
},
|
| 35 |
+
"image_size": 224,
|
| 36 |
+
"initializer_range": 0.02,
|
| 37 |
+
"label2id": {
|
| 38 |
+
"safe": 1,
|
| 39 |
+
"unlabeled": 0,
|
| 40 |
+
"unsafe": 2
|
| 41 |
+
},
|
| 42 |
+
"layer_norm_eps": 1e-06,
|
| 43 |
+
"mlp_ratios": [
|
| 44 |
+
4,
|
| 45 |
+
4,
|
| 46 |
+
4,
|
| 47 |
+
4
|
| 48 |
+
],
|
| 49 |
+
"model_type": "segformer",
|
| 50 |
+
"num_attention_heads": [
|
| 51 |
+
1,
|
| 52 |
+
2,
|
| 53 |
+
5,
|
| 54 |
+
8
|
| 55 |
+
],
|
| 56 |
+
"num_channels": 3,
|
| 57 |
+
"num_encoder_blocks": 4,
|
| 58 |
+
"patch_sizes": [
|
| 59 |
+
7,
|
| 60 |
+
3,
|
| 61 |
+
3,
|
| 62 |
+
3
|
| 63 |
+
],
|
| 64 |
+
"reshape_last_stage": true,
|
| 65 |
+
"semantic_loss_ignore_index": 255,
|
| 66 |
+
"sr_ratios": [
|
| 67 |
+
8,
|
| 68 |
+
4,
|
| 69 |
+
2,
|
| 70 |
+
1
|
| 71 |
+
],
|
| 72 |
+
"strides": [
|
| 73 |
+
4,
|
| 74 |
+
2,
|
| 75 |
+
2,
|
| 76 |
+
2
|
| 77 |
+
],
|
| 78 |
+
"torch_dtype": "float32",
|
| 79 |
+
"transformers_version": "4.30.2"
|
| 80 |
+
}
|
checkpoint-80/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3eeac63976056bd459dd869b4c32580d56656fef8340fc97a5ec2fb961084ce7
|
| 3 |
+
size 677455103
|
checkpoint-80/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e5a317ae83515840b71d6e336dc6051c9c7d2058ca428729a042378c2980f2b
|
| 3 |
+
size 338796489
|
checkpoint-80/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1cebdc4e00d4b10ce551a0c867229ef2d035f4b9467924cc945aa458e5947aa9
|
| 3 |
+
size 14575
|
checkpoint-80/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b739b417a088ffe39f50647c898b65416ca9f31232767b692819f35fa33c2ff
|
| 3 |
+
size 627
|
checkpoint-80/trainer_state.json
ADDED
|
@@ -0,0 +1,564 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7629496455192566,
|
| 3 |
+
"best_model_checkpoint": "/robodata/smodak/Projects/nspl/scripts/terrainseg/training/models/INTERNAL_BEST-safety-utcustom-train-SF-RGB-b5/checkpoint-80",
|
| 4 |
+
"epoch": 8.0,
|
| 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": 0.1,
|
| 12 |
+
"learning_rate": 1.0000000000000001e-07,
|
| 13 |
+
"loss": 1.2416,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.2,
|
| 18 |
+
"learning_rate": 2.0000000000000002e-07,
|
| 19 |
+
"loss": 1.2501,
|
| 20 |
+
"step": 2
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.3,
|
| 24 |
+
"learning_rate": 3.0000000000000004e-07,
|
| 25 |
+
"loss": 1.2448,
|
| 26 |
+
"step": 3
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.4,
|
| 30 |
+
"learning_rate": 4.0000000000000003e-07,
|
| 31 |
+
"loss": 1.2497,
|
| 32 |
+
"step": 4
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.5,
|
| 36 |
+
"learning_rate": 5.000000000000001e-07,
|
| 37 |
+
"loss": 1.2596,
|
| 38 |
+
"step": 5
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.6,
|
| 42 |
+
"learning_rate": 6.000000000000001e-07,
|
| 43 |
+
"loss": 1.2563,
|
| 44 |
+
"step": 6
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.7,
|
| 48 |
+
"learning_rate": 7.000000000000001e-07,
|
| 49 |
+
"loss": 1.2528,
|
| 50 |
+
"step": 7
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.8,
|
| 54 |
+
"learning_rate": 8.000000000000001e-07,
|
| 55 |
+
"loss": 1.2488,
|
| 56 |
+
"step": 8
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.9,
|
| 60 |
+
"learning_rate": 9e-07,
|
| 61 |
+
"loss": 1.246,
|
| 62 |
+
"step": 9
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.0,
|
| 66 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 67 |
+
"loss": 1.2569,
|
| 68 |
+
"step": 10
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.1,
|
| 72 |
+
"learning_rate": 1.1e-06,
|
| 73 |
+
"loss": 1.2506,
|
| 74 |
+
"step": 11
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.2,
|
| 78 |
+
"learning_rate": 1.2000000000000002e-06,
|
| 79 |
+
"loss": 1.2516,
|
| 80 |
+
"step": 12
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.3,
|
| 84 |
+
"learning_rate": 1.3e-06,
|
| 85 |
+
"loss": 1.2334,
|
| 86 |
+
"step": 13
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.4,
|
| 90 |
+
"learning_rate": 1.4000000000000001e-06,
|
| 91 |
+
"loss": 1.2405,
|
| 92 |
+
"step": 14
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.5,
|
| 96 |
+
"learning_rate": 1.5e-06,
|
| 97 |
+
"loss": 1.2353,
|
| 98 |
+
"step": 15
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.6,
|
| 102 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 103 |
+
"loss": 1.2381,
|
| 104 |
+
"step": 16
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.7,
|
| 108 |
+
"learning_rate": 1.7000000000000002e-06,
|
| 109 |
+
"loss": 1.2353,
|
| 110 |
+
"step": 17
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 1.8,
|
| 114 |
+
"learning_rate": 1.8e-06,
|
| 115 |
+
"loss": 1.2275,
|
| 116 |
+
"step": 18
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.9,
|
| 120 |
+
"learning_rate": 1.9e-06,
|
| 121 |
+
"loss": 1.2315,
|
| 122 |
+
"step": 19
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 2.0,
|
| 126 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 127 |
+
"loss": 1.2199,
|
| 128 |
+
"step": 20
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.0,
|
| 132 |
+
"eval_accuracy_safe": 0.6769557557603865,
|
| 133 |
+
"eval_accuracy_unlabeled": NaN,
|
| 134 |
+
"eval_accuracy_unsafe": 0.20277138320874238,
|
| 135 |
+
"eval_iou_safe": 0.027914850351298256,
|
| 136 |
+
"eval_iou_unlabeled": 0.0,
|
| 137 |
+
"eval_iou_unsafe": 0.20144342824057224,
|
| 138 |
+
"eval_loss": 1.147387981414795,
|
| 139 |
+
"eval_mean_accuracy": 0.4398635694845644,
|
| 140 |
+
"eval_mean_iou": 0.0764527595306235,
|
| 141 |
+
"eval_overall_accuracy": 0.2167780292567922,
|
| 142 |
+
"eval_runtime": 9.4386,
|
| 143 |
+
"eval_samples_per_second": 7.098,
|
| 144 |
+
"eval_steps_per_second": 0.53,
|
| 145 |
+
"step": 20
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 2.1,
|
| 149 |
+
"learning_rate": 2.1000000000000002e-06,
|
| 150 |
+
"loss": 1.2169,
|
| 151 |
+
"step": 21
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 2.2,
|
| 155 |
+
"learning_rate": 2.2e-06,
|
| 156 |
+
"loss": 1.2233,
|
| 157 |
+
"step": 22
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 2.3,
|
| 161 |
+
"learning_rate": 2.3e-06,
|
| 162 |
+
"loss": 1.225,
|
| 163 |
+
"step": 23
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 2.4,
|
| 167 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 168 |
+
"loss": 1.2193,
|
| 169 |
+
"step": 24
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 2.5,
|
| 173 |
+
"learning_rate": 2.5e-06,
|
| 174 |
+
"loss": 1.2153,
|
| 175 |
+
"step": 25
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 2.6,
|
| 179 |
+
"learning_rate": 2.6e-06,
|
| 180 |
+
"loss": 1.2085,
|
| 181 |
+
"step": 26
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.7,
|
| 185 |
+
"learning_rate": 2.7e-06,
|
| 186 |
+
"loss": 1.2061,
|
| 187 |
+
"step": 27
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 2.8,
|
| 191 |
+
"learning_rate": 2.8000000000000003e-06,
|
| 192 |
+
"loss": 1.2082,
|
| 193 |
+
"step": 28
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.9,
|
| 197 |
+
"learning_rate": 2.9e-06,
|
| 198 |
+
"loss": 1.208,
|
| 199 |
+
"step": 29
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 3.0,
|
| 203 |
+
"learning_rate": 3e-06,
|
| 204 |
+
"loss": 1.205,
|
| 205 |
+
"step": 30
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 3.1,
|
| 209 |
+
"learning_rate": 3.1e-06,
|
| 210 |
+
"loss": 1.1978,
|
| 211 |
+
"step": 31
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 3.2,
|
| 215 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 216 |
+
"loss": 1.1951,
|
| 217 |
+
"step": 32
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 3.3,
|
| 221 |
+
"learning_rate": 3.3e-06,
|
| 222 |
+
"loss": 1.1905,
|
| 223 |
+
"step": 33
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 3.4,
|
| 227 |
+
"learning_rate": 3.4000000000000005e-06,
|
| 228 |
+
"loss": 1.1892,
|
| 229 |
+
"step": 34
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 3.5,
|
| 233 |
+
"learning_rate": 3.5000000000000004e-06,
|
| 234 |
+
"loss": 1.1726,
|
| 235 |
+
"step": 35
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 3.6,
|
| 239 |
+
"learning_rate": 3.6e-06,
|
| 240 |
+
"loss": 1.1638,
|
| 241 |
+
"step": 36
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 3.7,
|
| 245 |
+
"learning_rate": 3.7e-06,
|
| 246 |
+
"loss": 1.1697,
|
| 247 |
+
"step": 37
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 3.8,
|
| 251 |
+
"learning_rate": 3.8e-06,
|
| 252 |
+
"loss": 1.1557,
|
| 253 |
+
"step": 38
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 3.9,
|
| 257 |
+
"learning_rate": 3.9e-06,
|
| 258 |
+
"loss": 1.1568,
|
| 259 |
+
"step": 39
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 4.0,
|
| 263 |
+
"learning_rate": 4.000000000000001e-06,
|
| 264 |
+
"loss": 1.1542,
|
| 265 |
+
"step": 40
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 4.0,
|
| 269 |
+
"eval_accuracy_safe": 0.7907563964672457,
|
| 270 |
+
"eval_accuracy_unlabeled": NaN,
|
| 271 |
+
"eval_accuracy_unsafe": 0.42571478791888173,
|
| 272 |
+
"eval_iou_safe": 0.04364337256292872,
|
| 273 |
+
"eval_iou_unlabeled": 0.0,
|
| 274 |
+
"eval_iou_unsafe": 0.4237482501114813,
|
| 275 |
+
"eval_loss": 1.0615942478179932,
|
| 276 |
+
"eval_mean_accuracy": 0.6082355921930638,
|
| 277 |
+
"eval_mean_iou": 0.15579720755813667,
|
| 278 |
+
"eval_overall_accuracy": 0.4364975317200618,
|
| 279 |
+
"eval_runtime": 9.575,
|
| 280 |
+
"eval_samples_per_second": 6.997,
|
| 281 |
+
"eval_steps_per_second": 0.522,
|
| 282 |
+
"step": 40
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 4.1,
|
| 286 |
+
"learning_rate": 4.1000000000000006e-06,
|
| 287 |
+
"loss": 1.1516,
|
| 288 |
+
"step": 41
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 4.2,
|
| 292 |
+
"learning_rate": 4.2000000000000004e-06,
|
| 293 |
+
"loss": 1.1479,
|
| 294 |
+
"step": 42
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 4.3,
|
| 298 |
+
"learning_rate": 4.2999999999999995e-06,
|
| 299 |
+
"loss": 1.1341,
|
| 300 |
+
"step": 43
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 4.4,
|
| 304 |
+
"learning_rate": 4.4e-06,
|
| 305 |
+
"loss": 1.1384,
|
| 306 |
+
"step": 44
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 4.5,
|
| 310 |
+
"learning_rate": 4.5e-06,
|
| 311 |
+
"loss": 1.1202,
|
| 312 |
+
"step": 45
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 4.6,
|
| 316 |
+
"learning_rate": 4.6e-06,
|
| 317 |
+
"loss": 1.1091,
|
| 318 |
+
"step": 46
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 4.7,
|
| 322 |
+
"learning_rate": 4.7e-06,
|
| 323 |
+
"loss": 1.112,
|
| 324 |
+
"step": 47
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 4.8,
|
| 328 |
+
"learning_rate": 4.800000000000001e-06,
|
| 329 |
+
"loss": 1.1014,
|
| 330 |
+
"step": 48
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 4.9,
|
| 334 |
+
"learning_rate": 4.9000000000000005e-06,
|
| 335 |
+
"loss": 1.1065,
|
| 336 |
+
"step": 49
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 5.0,
|
| 340 |
+
"learning_rate": 5e-06,
|
| 341 |
+
"loss": 1.1028,
|
| 342 |
+
"step": 50
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 5.1,
|
| 346 |
+
"learning_rate": 5.1e-06,
|
| 347 |
+
"loss": 1.0823,
|
| 348 |
+
"step": 51
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 5.2,
|
| 352 |
+
"learning_rate": 5.2e-06,
|
| 353 |
+
"loss": 1.0721,
|
| 354 |
+
"step": 52
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 5.3,
|
| 358 |
+
"learning_rate": 5.3e-06,
|
| 359 |
+
"loss": 1.0809,
|
| 360 |
+
"step": 53
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 5.4,
|
| 364 |
+
"learning_rate": 5.4e-06,
|
| 365 |
+
"loss": 1.0711,
|
| 366 |
+
"step": 54
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 5.5,
|
| 370 |
+
"learning_rate": 5.500000000000001e-06,
|
| 371 |
+
"loss": 1.0566,
|
| 372 |
+
"step": 55
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 5.6,
|
| 376 |
+
"learning_rate": 5.600000000000001e-06,
|
| 377 |
+
"loss": 1.0401,
|
| 378 |
+
"step": 56
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 5.7,
|
| 382 |
+
"learning_rate": 5.7000000000000005e-06,
|
| 383 |
+
"loss": 1.073,
|
| 384 |
+
"step": 57
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 5.8,
|
| 388 |
+
"learning_rate": 5.8e-06,
|
| 389 |
+
"loss": 1.0281,
|
| 390 |
+
"step": 58
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 5.9,
|
| 394 |
+
"learning_rate": 5.9e-06,
|
| 395 |
+
"loss": 1.0327,
|
| 396 |
+
"step": 59
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 6.0,
|
| 400 |
+
"learning_rate": 6e-06,
|
| 401 |
+
"loss": 1.0324,
|
| 402 |
+
"step": 60
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 6.0,
|
| 406 |
+
"eval_accuracy_safe": 0.6667148546073454,
|
| 407 |
+
"eval_accuracy_unlabeled": NaN,
|
| 408 |
+
"eval_accuracy_unsafe": 0.7104100559195431,
|
| 409 |
+
"eval_iou_safe": 0.06715586533607736,
|
| 410 |
+
"eval_iou_unlabeled": 0.0,
|
| 411 |
+
"eval_iou_unsafe": 0.7035684769007364,
|
| 412 |
+
"eval_loss": 0.9139688611030579,
|
| 413 |
+
"eval_mean_accuracy": 0.6885624552634442,
|
| 414 |
+
"eval_mean_iou": 0.25690811407893793,
|
| 415 |
+
"eval_overall_accuracy": 0.709119369734579,
|
| 416 |
+
"eval_runtime": 9.3129,
|
| 417 |
+
"eval_samples_per_second": 7.194,
|
| 418 |
+
"eval_steps_per_second": 0.537,
|
| 419 |
+
"step": 60
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 6.1,
|
| 423 |
+
"learning_rate": 6.1e-06,
|
| 424 |
+
"loss": 0.996,
|
| 425 |
+
"step": 61
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 6.2,
|
| 429 |
+
"learning_rate": 6.2e-06,
|
| 430 |
+
"loss": 1.0074,
|
| 431 |
+
"step": 62
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 6.3,
|
| 435 |
+
"learning_rate": 6.300000000000001e-06,
|
| 436 |
+
"loss": 1.0368,
|
| 437 |
+
"step": 63
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 6.4,
|
| 441 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 442 |
+
"loss": 0.981,
|
| 443 |
+
"step": 64
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 6.5,
|
| 447 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 448 |
+
"loss": 0.9959,
|
| 449 |
+
"step": 65
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 6.6,
|
| 453 |
+
"learning_rate": 6.6e-06,
|
| 454 |
+
"loss": 0.9577,
|
| 455 |
+
"step": 66
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 6.7,
|
| 459 |
+
"learning_rate": 6.700000000000001e-06,
|
| 460 |
+
"loss": 0.9664,
|
| 461 |
+
"step": 67
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 6.8,
|
| 465 |
+
"learning_rate": 6.800000000000001e-06,
|
| 466 |
+
"loss": 0.9371,
|
| 467 |
+
"step": 68
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 6.9,
|
| 471 |
+
"learning_rate": 6.900000000000001e-06,
|
| 472 |
+
"loss": 0.9334,
|
| 473 |
+
"step": 69
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 7.0,
|
| 477 |
+
"learning_rate": 7.000000000000001e-06,
|
| 478 |
+
"loss": 0.9258,
|
| 479 |
+
"step": 70
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 7.1,
|
| 483 |
+
"learning_rate": 7.1e-06,
|
| 484 |
+
"loss": 0.9073,
|
| 485 |
+
"step": 71
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 7.2,
|
| 489 |
+
"learning_rate": 7.2e-06,
|
| 490 |
+
"loss": 1.004,
|
| 491 |
+
"step": 72
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 7.3,
|
| 495 |
+
"learning_rate": 7.2999999999999996e-06,
|
| 496 |
+
"loss": 0.9029,
|
| 497 |
+
"step": 73
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 7.4,
|
| 501 |
+
"learning_rate": 7.4e-06,
|
| 502 |
+
"loss": 0.8715,
|
| 503 |
+
"step": 74
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 7.5,
|
| 507 |
+
"learning_rate": 7.5e-06,
|
| 508 |
+
"loss": 0.8641,
|
| 509 |
+
"step": 75
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 7.6,
|
| 513 |
+
"learning_rate": 7.6e-06,
|
| 514 |
+
"loss": 0.8934,
|
| 515 |
+
"step": 76
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 7.7,
|
| 519 |
+
"learning_rate": 7.7e-06,
|
| 520 |
+
"loss": 0.8487,
|
| 521 |
+
"step": 77
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 7.8,
|
| 525 |
+
"learning_rate": 7.8e-06,
|
| 526 |
+
"loss": 0.8626,
|
| 527 |
+
"step": 78
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 7.9,
|
| 531 |
+
"learning_rate": 7.9e-06,
|
| 532 |
+
"loss": 0.8309,
|
| 533 |
+
"step": 79
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 8.0,
|
| 537 |
+
"learning_rate": 8.000000000000001e-06,
|
| 538 |
+
"loss": 0.8058,
|
| 539 |
+
"step": 80
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 8.0,
|
| 543 |
+
"eval_accuracy_safe": 0.5359944641693748,
|
| 544 |
+
"eval_accuracy_unlabeled": NaN,
|
| 545 |
+
"eval_accuracy_unsafe": 0.8199187015242027,
|
| 546 |
+
"eval_iou_safe": 0.08228570650745816,
|
| 547 |
+
"eval_iou_unlabeled": 0.0,
|
| 548 |
+
"eval_iou_unsafe": 0.8086088911065791,
|
| 549 |
+
"eval_loss": 0.7629496455192566,
|
| 550 |
+
"eval_mean_accuracy": 0.6779565828467887,
|
| 551 |
+
"eval_mean_iou": 0.29696486587134574,
|
| 552 |
+
"eval_overall_accuracy": 0.8115320348027927,
|
| 553 |
+
"eval_runtime": 10.0198,
|
| 554 |
+
"eval_samples_per_second": 6.687,
|
| 555 |
+
"eval_steps_per_second": 0.499,
|
| 556 |
+
"step": 80
|
| 557 |
+
}
|
| 558 |
+
],
|
| 559 |
+
"max_steps": 20000,
|
| 560 |
+
"num_train_epochs": 2000,
|
| 561 |
+
"total_flos": 4.9178035003745894e+17,
|
| 562 |
+
"trial_name": null,
|
| 563 |
+
"trial_params": null
|
| 564 |
+
}
|
checkpoint-80/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2df02a3a55eecab0cc6613943484a0f9aa4844d2d4254173575c151fe91e8ce1
|
| 3 |
+
size 4219
|
config.json
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "nvidia/mit-b5",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"SegformerForSemanticSegmentation"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"classifier_dropout_prob": 0.1,
|
| 8 |
+
"decoder_hidden_size": 768,
|
| 9 |
+
"depths": [
|
| 10 |
+
3,
|
| 11 |
+
6,
|
| 12 |
+
40,
|
| 13 |
+
3
|
| 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 |
+
64,
|
| 26 |
+
128,
|
| 27 |
+
320,
|
| 28 |
+
512
|
| 29 |
+
],
|
| 30 |
+
"id2label": {
|
| 31 |
+
"0": "unlabeled",
|
| 32 |
+
"1": "safe",
|
| 33 |
+
"2": "unsafe"
|
| 34 |
+
},
|
| 35 |
+
"image_size": 224,
|
| 36 |
+
"initializer_range": 0.02,
|
| 37 |
+
"label2id": {
|
| 38 |
+
"safe": 1,
|
| 39 |
+
"unlabeled": 0,
|
| 40 |
+
"unsafe": 2
|
| 41 |
+
},
|
| 42 |
+
"layer_norm_eps": 1e-06,
|
| 43 |
+
"mlp_ratios": [
|
| 44 |
+
4,
|
| 45 |
+
4,
|
| 46 |
+
4,
|
| 47 |
+
4
|
| 48 |
+
],
|
| 49 |
+
"model_type": "segformer",
|
| 50 |
+
"num_attention_heads": [
|
| 51 |
+
1,
|
| 52 |
+
2,
|
| 53 |
+
5,
|
| 54 |
+
8
|
| 55 |
+
],
|
| 56 |
+
"num_channels": 3,
|
| 57 |
+
"num_encoder_blocks": 4,
|
| 58 |
+
"patch_sizes": [
|
| 59 |
+
7,
|
| 60 |
+
3,
|
| 61 |
+
3,
|
| 62 |
+
3
|
| 63 |
+
],
|
| 64 |
+
"reshape_last_stage": true,
|
| 65 |
+
"semantic_loss_ignore_index": 255,
|
| 66 |
+
"sr_ratios": [
|
| 67 |
+
8,
|
| 68 |
+
4,
|
| 69 |
+
2,
|
| 70 |
+
1
|
| 71 |
+
],
|
| 72 |
+
"strides": [
|
| 73 |
+
4,
|
| 74 |
+
2,
|
| 75 |
+
2,
|
| 76 |
+
2
|
| 77 |
+
],
|
| 78 |
+
"torch_dtype": "float32",
|
| 79 |
+
"transformers_version": "4.30.2"
|
| 80 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e5a317ae83515840b71d6e336dc6051c9c7d2058ca428729a042378c2980f2b
|
| 3 |
+
size 338796489
|
runs/Feb08_09-49-29_robovision.csres.utexas.edu/events.out.tfevents.1707407376.robovision.csres.utexas.edu.889113.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:614a7e5cf8025ad64980e3328e42be48f8fe6ce3d8539a0178687f96b0939b67
|
| 3 |
+
size 20198
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2df02a3a55eecab0cc6613943484a0f9aa4844d2d4254173575c151fe91e8ce1
|
| 3 |
+
size 4219
|