End of training
Browse files
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.5419354838709678,
|
| 4 |
+
"eval_loss": 1.295546531677246,
|
| 5 |
+
"eval_runtime": 17.0543,
|
| 6 |
+
"eval_samples_per_second": 9.089,
|
| 7 |
+
"eval_steps_per_second": 9.089
|
| 8 |
}
|
runs/Sep24_15-21-40_stupa-ai/events.out.tfevents.1727191488.stupa-ai.2487498.1
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4a05dd766f7022e833f23bf2fcda291adc33048beb137d6096723ad1ded263a9
|
| 3 |
+
size 734
|
test_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second":
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_accuracy": 0.5419354838709678,
|
| 4 |
+
"eval_loss": 1.295546531677246,
|
| 5 |
+
"eval_runtime": 17.0543,
|
| 6 |
+
"eval_samples_per_second": 9.089,
|
| 7 |
+
"eval_steps_per_second": 9.089
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,107 +1,262 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/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": 0.
|
| 13 |
-
"grad_norm":
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss":
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"
|
| 21 |
-
"
|
| 22 |
-
"
|
| 23 |
-
"
|
| 24 |
-
"eval_steps_per_second": 0.246,
|
| 25 |
-
"step": 12
|
| 26 |
},
|
| 27 |
{
|
| 28 |
-
"epoch": 1
|
| 29 |
-
"grad_norm":
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss":
|
| 32 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch":
|
| 36 |
-
"
|
| 37 |
-
"
|
| 38 |
-
"
|
| 39 |
-
"
|
| 40 |
-
"eval_steps_per_second": 0.265,
|
| 41 |
-
"step": 24
|
| 42 |
},
|
| 43 |
{
|
| 44 |
-
"epoch": 2
|
| 45 |
-
"grad_norm":
|
| 46 |
-
"learning_rate":
|
| 47 |
-
"loss":
|
| 48 |
-
"step":
|
| 49 |
},
|
| 50 |
{
|
| 51 |
-
"epoch":
|
| 52 |
-
"
|
| 53 |
-
"
|
| 54 |
-
"
|
| 55 |
-
"
|
| 56 |
-
"eval_steps_per_second": 0.265,
|
| 57 |
-
"step": 36
|
| 58 |
},
|
| 59 |
{
|
| 60 |
-
"epoch":
|
| 61 |
-
"grad_norm": 20.
|
| 62 |
-
"learning_rate":
|
| 63 |
-
"loss":
|
| 64 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 65 |
},
|
| 66 |
{
|
| 67 |
-
"epoch":
|
| 68 |
-
"eval_accuracy": 0.
|
| 69 |
-
"eval_loss":
|
| 70 |
-
"eval_runtime":
|
| 71 |
-
"eval_samples_per_second":
|
| 72 |
-
"eval_steps_per_second":
|
| 73 |
-
"step":
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"epoch": 3.25,
|
| 77 |
-
"step": 48,
|
| 78 |
-
"total_flos": 5.981110029464371e+16,
|
| 79 |
-
"train_loss": 0.7539124091466268,
|
| 80 |
-
"train_runtime": 598.1535,
|
| 81 |
-
"train_samples_per_second": 0.08,
|
| 82 |
-
"train_steps_per_second": 0.08
|
| 83 |
-
},
|
| 84 |
-
{
|
| 85 |
-
"epoch": 3.25,
|
| 86 |
-
"eval_accuracy": 0.75,
|
| 87 |
-
"eval_loss": 0.6049277186393738,
|
| 88 |
-
"eval_runtime": 51.052,
|
| 89 |
-
"eval_samples_per_second": 0.235,
|
| 90 |
-
"eval_steps_per_second": 0.235,
|
| 91 |
-
"step": 48
|
| 92 |
-
},
|
| 93 |
-
{
|
| 94 |
-
"epoch": 3.25,
|
| 95 |
-
"eval_accuracy": 0.75,
|
| 96 |
-
"eval_loss": 0.6049277782440186,
|
| 97 |
-
"eval_runtime": 45.5299,
|
| 98 |
-
"eval_samples_per_second": 0.264,
|
| 99 |
-
"eval_steps_per_second": 0.264,
|
| 100 |
-
"step": 48
|
| 101 |
}
|
| 102 |
],
|
| 103 |
"logging_steps": 10,
|
| 104 |
-
"max_steps":
|
| 105 |
"num_input_tokens_seen": 0,
|
| 106 |
"num_train_epochs": 9223372036854775807,
|
| 107 |
"save_steps": 500,
|
|
@@ -117,7 +272,7 @@
|
|
| 117 |
"attributes": {}
|
| 118 |
}
|
| 119 |
},
|
| 120 |
-
"total_flos":
|
| 121 |
"train_batch_size": 1,
|
| 122 |
"trial_name": null,
|
| 123 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.6428571428571429,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-300",
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 300,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.03333333333333333,
|
| 13 |
+
"grad_norm": 37.11217498779297,
|
| 14 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 15 |
+
"loss": 2.256,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.06666666666666667,
|
| 20 |
+
"grad_norm": 20.375593185424805,
|
| 21 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 22 |
+
"loss": 2.2086,
|
| 23 |
+
"step": 20
|
|
|
|
|
|
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.1,
|
| 27 |
+
"grad_norm": 21.34927749633789,
|
| 28 |
+
"learning_rate": 5e-05,
|
| 29 |
+
"loss": 2.628,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.13333333333333333,
|
| 34 |
+
"grad_norm": 18.237977981567383,
|
| 35 |
+
"learning_rate": 4.814814814814815e-05,
|
| 36 |
+
"loss": 2.4005,
|
| 37 |
+
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.16666666666666666,
|
| 41 |
+
"grad_norm": 18.60643768310547,
|
| 42 |
+
"learning_rate": 4.62962962962963e-05,
|
| 43 |
+
"loss": 2.374,
|
| 44 |
+
"step": 50
|
|
|
|
|
|
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.2,
|
| 48 |
+
"grad_norm": 19.271390914916992,
|
| 49 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 50 |
+
"loss": 2.2412,
|
| 51 |
+
"step": 60
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.23333333333333334,
|
| 55 |
+
"grad_norm": 20.931049346923828,
|
| 56 |
+
"learning_rate": 4.259259259259259e-05,
|
| 57 |
+
"loss": 2.0993,
|
| 58 |
+
"step": 70
|
|
|
|
|
|
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.26666666666666666,
|
| 62 |
+
"grad_norm": 20.036760330200195,
|
| 63 |
+
"learning_rate": 4.074074074074074e-05,
|
| 64 |
+
"loss": 2.4346,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.3,
|
| 69 |
+
"grad_norm": 19.907039642333984,
|
| 70 |
+
"learning_rate": 3.888888888888889e-05,
|
| 71 |
+
"loss": 1.9128,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.3333333333333333,
|
| 76 |
+
"grad_norm": 22.826419830322266,
|
| 77 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 78 |
+
"loss": 1.8796,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.36666666666666664,
|
| 83 |
+
"grad_norm": 27.385215759277344,
|
| 84 |
+
"learning_rate": 3.518518518518519e-05,
|
| 85 |
+
"loss": 2.1521,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.4,
|
| 90 |
+
"grad_norm": 22.632038116455078,
|
| 91 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 92 |
+
"loss": 2.115,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.43333333333333335,
|
| 97 |
+
"grad_norm": 25.79122543334961,
|
| 98 |
+
"learning_rate": 3.148148148148148e-05,
|
| 99 |
+
"loss": 2.316,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.4666666666666667,
|
| 104 |
+
"grad_norm": 40.01160430908203,
|
| 105 |
+
"learning_rate": 2.962962962962963e-05,
|
| 106 |
+
"loss": 1.8187,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.5,
|
| 111 |
+
"grad_norm": 16.484312057495117,
|
| 112 |
+
"learning_rate": 2.777777777777778e-05,
|
| 113 |
+
"loss": 1.7254,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.5333333333333333,
|
| 118 |
+
"grad_norm": 26.357309341430664,
|
| 119 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 120 |
+
"loss": 1.6207,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.5666666666666667,
|
| 125 |
+
"grad_norm": 20.33473777770996,
|
| 126 |
+
"learning_rate": 2.4074074074074074e-05,
|
| 127 |
+
"loss": 1.5019,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.6,
|
| 132 |
+
"grad_norm": 23.332971572875977,
|
| 133 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 134 |
+
"loss": 1.8112,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.6333333333333333,
|
| 139 |
+
"grad_norm": 17.828908920288086,
|
| 140 |
+
"learning_rate": 2.037037037037037e-05,
|
| 141 |
+
"loss": 1.4698,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.6666666666666666,
|
| 146 |
+
"grad_norm": 15.446304321289062,
|
| 147 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 148 |
+
"loss": 1.4156,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.7,
|
| 153 |
+
"grad_norm": 13.840889930725098,
|
| 154 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 155 |
+
"loss": 2.161,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.7333333333333333,
|
| 160 |
+
"grad_norm": 20.35426902770996,
|
| 161 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 162 |
+
"loss": 1.0496,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.7666666666666667,
|
| 167 |
+
"grad_norm": 10.623125076293945,
|
| 168 |
+
"learning_rate": 1.2962962962962962e-05,
|
| 169 |
+
"loss": 1.2846,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.8,
|
| 174 |
+
"grad_norm": 16.291202545166016,
|
| 175 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 176 |
+
"loss": 1.5562,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.8333333333333334,
|
| 181 |
+
"grad_norm": 28.847074508666992,
|
| 182 |
+
"learning_rate": 9.259259259259259e-06,
|
| 183 |
+
"loss": 1.3731,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.8666666666666667,
|
| 188 |
+
"grad_norm": 21.897621154785156,
|
| 189 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 190 |
+
"loss": 1.8421,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.9,
|
| 195 |
+
"grad_norm": 16.28655242919922,
|
| 196 |
+
"learning_rate": 5.555555555555556e-06,
|
| 197 |
+
"loss": 1.4873,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.9333333333333333,
|
| 202 |
+
"grad_norm": 12.424924850463867,
|
| 203 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 204 |
+
"loss": 1.4756,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.9666666666666667,
|
| 209 |
+
"grad_norm": 20.726268768310547,
|
| 210 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 211 |
+
"loss": 0.9627,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.0,
|
| 216 |
+
"grad_norm": 24.575002670288086,
|
| 217 |
+
"learning_rate": 0.0,
|
| 218 |
+
"loss": 1.2724,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.0,
|
| 223 |
+
"eval_accuracy": 0.6428571428571429,
|
| 224 |
+
"eval_loss": 1.1614660024642944,
|
| 225 |
+
"eval_runtime": 7.7497,
|
| 226 |
+
"eval_samples_per_second": 9.033,
|
| 227 |
+
"eval_steps_per_second": 9.033,
|
| 228 |
+
"step": 300
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 1.0,
|
| 232 |
+
"step": 300,
|
| 233 |
+
"total_flos": 3.73846047031296e+17,
|
| 234 |
+
"train_loss": 1.8281906827290852,
|
| 235 |
+
"train_runtime": 79.9931,
|
| 236 |
+
"train_samples_per_second": 3.75,
|
| 237 |
+
"train_steps_per_second": 3.75
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.0,
|
| 241 |
+
"eval_accuracy": 0.5419354838709678,
|
| 242 |
+
"eval_loss": 1.295546531677246,
|
| 243 |
+
"eval_runtime": 17.2162,
|
| 244 |
+
"eval_samples_per_second": 9.003,
|
| 245 |
+
"eval_steps_per_second": 9.003,
|
| 246 |
+
"step": 300
|
| 247 |
},
|
| 248 |
{
|
| 249 |
+
"epoch": 1.0,
|
| 250 |
+
"eval_accuracy": 0.5419354838709678,
|
| 251 |
+
"eval_loss": 1.295546531677246,
|
| 252 |
+
"eval_runtime": 17.0543,
|
| 253 |
+
"eval_samples_per_second": 9.089,
|
| 254 |
+
"eval_steps_per_second": 9.089,
|
| 255 |
+
"step": 300
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 256 |
}
|
| 257 |
],
|
| 258 |
"logging_steps": 10,
|
| 259 |
+
"max_steps": 300,
|
| 260 |
"num_input_tokens_seen": 0,
|
| 261 |
"num_train_epochs": 9223372036854775807,
|
| 262 |
"save_steps": 500,
|
|
|
|
| 272 |
"attributes": {}
|
| 273 |
}
|
| 274 |
},
|
| 275 |
+
"total_flos": 3.73846047031296e+17,
|
| 276 |
"train_batch_size": 1,
|
| 277 |
"trial_name": null,
|
| 278 |
"trial_params": null
|