🍻 cheers
Browse files- README.md +2 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Apr20_09-54-21_c768f2f23310/events.out.tfevents.1713640272.c768f2f23310.228.5 +3 -0
- train_results.json +8 -0
- trainer_state.json +400 -0
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/convnext-tiny-224
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# idbwtiny
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [facebook/convnext-tiny-224](https://huggingface.co/facebook/convnext-tiny-224) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.1267
|
| 21 |
- Accuracy: 0.9569
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/convnext-tiny-224
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
|
|
| 16 |
|
| 17 |
# idbwtiny
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [facebook/convnext-tiny-224](https://huggingface.co/facebook/convnext-tiny-224) on the indian_food_images dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.1267
|
| 22 |
- Accuracy: 0.9569
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9568830973467968,
|
| 4 |
+
"eval_loss": 0.12666618824005127,
|
| 5 |
+
"eval_runtime": 203.8079,
|
| 6 |
+
"eval_samples_per_second": 260.937,
|
| 7 |
+
"eval_steps_per_second": 32.619,
|
| 8 |
+
"total_flos": 7.572439992003492e+19,
|
| 9 |
+
"train_loss": 0.2540779383265557,
|
| 10 |
+
"train_runtime": 33188.7721,
|
| 11 |
+
"train_samples_per_second": 90.801,
|
| 12 |
+
"train_steps_per_second": 0.606
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9568830973467968,
|
| 4 |
+
"eval_loss": 0.12666618824005127,
|
| 5 |
+
"eval_runtime": 203.8079,
|
| 6 |
+
"eval_samples_per_second": 260.937,
|
| 7 |
+
"eval_steps_per_second": 32.619
|
| 8 |
+
}
|
runs/Apr20_09-54-21_c768f2f23310/events.out.tfevents.1713640272.c768f2f23310.228.5
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb8bfb9c31b5088d98982b7fd7959b4789462202c9b8ced043dd9d9db12f5bdb
|
| 3 |
+
size 417
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 7.572439992003492e+19,
|
| 4 |
+
"train_loss": 0.2540779383265557,
|
| 5 |
+
"train_runtime": 33188.7721,
|
| 6 |
+
"train_samples_per_second": 90.801,
|
| 7 |
+
"train_steps_per_second": 0.606
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,400 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.12666618824005127,
|
| 3 |
+
"best_model_checkpoint": "idbwtiny/checkpoint-20100",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 20100,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.25,
|
| 13 |
+
"grad_norm": 2.5227627754211426,
|
| 14 |
+
"learning_rate": 0.00019503482587064676,
|
| 15 |
+
"loss": 0.4184,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.5,
|
| 20 |
+
"grad_norm": 2.693358898162842,
|
| 21 |
+
"learning_rate": 0.0001900597014925373,
|
| 22 |
+
"loss": 0.4215,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.75,
|
| 27 |
+
"grad_norm": 1.985125184059143,
|
| 28 |
+
"learning_rate": 0.00018508457711442785,
|
| 29 |
+
"loss": 0.3987,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.0,
|
| 34 |
+
"grad_norm": 1.946892261505127,
|
| 35 |
+
"learning_rate": 0.0001801094527363184,
|
| 36 |
+
"loss": 0.3871,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_accuracy": 0.8674150542486978,
|
| 42 |
+
"eval_loss": 0.3020988404750824,
|
| 43 |
+
"eval_runtime": 206.2751,
|
| 44 |
+
"eval_samples_per_second": 257.816,
|
| 45 |
+
"eval_steps_per_second": 32.229,
|
| 46 |
+
"step": 2010
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.24,
|
| 50 |
+
"grad_norm": 2.351902723312378,
|
| 51 |
+
"learning_rate": 0.00017514427860696518,
|
| 52 |
+
"loss": 0.3685,
|
| 53 |
+
"step": 2500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.49,
|
| 57 |
+
"grad_norm": 2.192362070083618,
|
| 58 |
+
"learning_rate": 0.00017016915422885572,
|
| 59 |
+
"loss": 0.3615,
|
| 60 |
+
"step": 3000
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.74,
|
| 64 |
+
"grad_norm": 2.392155647277832,
|
| 65 |
+
"learning_rate": 0.00016519402985074627,
|
| 66 |
+
"loss": 0.3474,
|
| 67 |
+
"step": 3500
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.99,
|
| 71 |
+
"grad_norm": 2.2455360889434814,
|
| 72 |
+
"learning_rate": 0.00016021890547263681,
|
| 73 |
+
"loss": 0.3403,
|
| 74 |
+
"step": 4000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.0,
|
| 78 |
+
"eval_accuracy": 0.8966548203305692,
|
| 79 |
+
"eval_loss": 0.24423933029174805,
|
| 80 |
+
"eval_runtime": 204.2994,
|
| 81 |
+
"eval_samples_per_second": 260.309,
|
| 82 |
+
"eval_steps_per_second": 32.54,
|
| 83 |
+
"step": 4020
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.24,
|
| 87 |
+
"grad_norm": 2.0239338874816895,
|
| 88 |
+
"learning_rate": 0.0001552437810945274,
|
| 89 |
+
"loss": 0.3266,
|
| 90 |
+
"step": 4500
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.49,
|
| 94 |
+
"grad_norm": 2.220857858657837,
|
| 95 |
+
"learning_rate": 0.00015027860696517414,
|
| 96 |
+
"loss": 0.316,
|
| 97 |
+
"step": 5000
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.74,
|
| 101 |
+
"grad_norm": 1.665713906288147,
|
| 102 |
+
"learning_rate": 0.00014530348258706468,
|
| 103 |
+
"loss": 0.3089,
|
| 104 |
+
"step": 5500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.99,
|
| 108 |
+
"grad_norm": 2.8135392665863037,
|
| 109 |
+
"learning_rate": 0.00014032835820895523,
|
| 110 |
+
"loss": 0.3058,
|
| 111 |
+
"step": 6000
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 3.0,
|
| 115 |
+
"eval_accuracy": 0.9052105075120814,
|
| 116 |
+
"eval_loss": 0.2269268035888672,
|
| 117 |
+
"eval_runtime": 209.4743,
|
| 118 |
+
"eval_samples_per_second": 253.878,
|
| 119 |
+
"eval_steps_per_second": 31.737,
|
| 120 |
+
"step": 6030
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.23,
|
| 124 |
+
"grad_norm": 2.630223512649536,
|
| 125 |
+
"learning_rate": 0.000135363184079602,
|
| 126 |
+
"loss": 0.2868,
|
| 127 |
+
"step": 6500
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.48,
|
| 131 |
+
"grad_norm": 1.8844317197799683,
|
| 132 |
+
"learning_rate": 0.00013038805970149256,
|
| 133 |
+
"loss": 0.2843,
|
| 134 |
+
"step": 7000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.73,
|
| 138 |
+
"grad_norm": 2.541440010070801,
|
| 139 |
+
"learning_rate": 0.0001254129353233831,
|
| 140 |
+
"loss": 0.2794,
|
| 141 |
+
"step": 7500
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.98,
|
| 145 |
+
"grad_norm": 2.5618529319763184,
|
| 146 |
+
"learning_rate": 0.00012043781094527365,
|
| 147 |
+
"loss": 0.2722,
|
| 148 |
+
"step": 8000
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 4.0,
|
| 152 |
+
"eval_accuracy": 0.9196893627423328,
|
| 153 |
+
"eval_loss": 0.20079384744167328,
|
| 154 |
+
"eval_runtime": 208.4042,
|
| 155 |
+
"eval_samples_per_second": 255.182,
|
| 156 |
+
"eval_steps_per_second": 31.9,
|
| 157 |
+
"step": 8040
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 4.23,
|
| 161 |
+
"grad_norm": 1.9443061351776123,
|
| 162 |
+
"learning_rate": 0.0001154726368159204,
|
| 163 |
+
"loss": 0.2546,
|
| 164 |
+
"step": 8500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.48,
|
| 168 |
+
"grad_norm": 1.7675588130950928,
|
| 169 |
+
"learning_rate": 0.00011049751243781094,
|
| 170 |
+
"loss": 0.2498,
|
| 171 |
+
"step": 9000
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 4.73,
|
| 175 |
+
"grad_norm": 2.6675548553466797,
|
| 176 |
+
"learning_rate": 0.00010552238805970149,
|
| 177 |
+
"loss": 0.2472,
|
| 178 |
+
"step": 9500
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.98,
|
| 182 |
+
"grad_norm": 3.0125935077667236,
|
| 183 |
+
"learning_rate": 0.00010054726368159203,
|
| 184 |
+
"loss": 0.2496,
|
| 185 |
+
"step": 10000
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 5.0,
|
| 189 |
+
"eval_accuracy": 0.9271920422707358,
|
| 190 |
+
"eval_loss": 0.184106707572937,
|
| 191 |
+
"eval_runtime": 208.5204,
|
| 192 |
+
"eval_samples_per_second": 255.04,
|
| 193 |
+
"eval_steps_per_second": 31.882,
|
| 194 |
+
"step": 10050
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 5.22,
|
| 198 |
+
"grad_norm": 3.092007637023926,
|
| 199 |
+
"learning_rate": 9.55721393034826e-05,
|
| 200 |
+
"loss": 0.2271,
|
| 201 |
+
"step": 10500
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 5.47,
|
| 205 |
+
"grad_norm": 4.371829509735107,
|
| 206 |
+
"learning_rate": 9.059701492537314e-05,
|
| 207 |
+
"loss": 0.2264,
|
| 208 |
+
"step": 11000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 5.72,
|
| 212 |
+
"grad_norm": 2.334900140762329,
|
| 213 |
+
"learning_rate": 8.562189054726368e-05,
|
| 214 |
+
"loss": 0.2257,
|
| 215 |
+
"step": 11500
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 5.97,
|
| 219 |
+
"grad_norm": 3.219524621963501,
|
| 220 |
+
"learning_rate": 8.064676616915423e-05,
|
| 221 |
+
"loss": 0.2219,
|
| 222 |
+
"step": 12000
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 6.0,
|
| 226 |
+
"eval_accuracy": 0.9352588330418758,
|
| 227 |
+
"eval_loss": 0.1706165224313736,
|
| 228 |
+
"eval_runtime": 212.161,
|
| 229 |
+
"eval_samples_per_second": 250.663,
|
| 230 |
+
"eval_steps_per_second": 31.335,
|
| 231 |
+
"step": 12060
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 6.22,
|
| 235 |
+
"grad_norm": 4.32515287399292,
|
| 236 |
+
"learning_rate": 7.5681592039801e-05,
|
| 237 |
+
"loss": 0.207,
|
| 238 |
+
"step": 12500
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 6.47,
|
| 242 |
+
"grad_norm": 2.8301429748535156,
|
| 243 |
+
"learning_rate": 7.070646766169154e-05,
|
| 244 |
+
"loss": 0.1996,
|
| 245 |
+
"step": 13000
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 6.72,
|
| 249 |
+
"grad_norm": 4.194028377532959,
|
| 250 |
+
"learning_rate": 6.573134328358209e-05,
|
| 251 |
+
"loss": 0.2018,
|
| 252 |
+
"step": 13500
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 6.97,
|
| 256 |
+
"grad_norm": 3.551224708557129,
|
| 257 |
+
"learning_rate": 6.075621890547264e-05,
|
| 258 |
+
"loss": 0.198,
|
| 259 |
+
"step": 14000
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 7.0,
|
| 263 |
+
"eval_accuracy": 0.9451871909140482,
|
| 264 |
+
"eval_loss": 0.147793248295784,
|
| 265 |
+
"eval_runtime": 210.5297,
|
| 266 |
+
"eval_samples_per_second": 252.606,
|
| 267 |
+
"eval_steps_per_second": 31.577,
|
| 268 |
+
"step": 14070
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 7.21,
|
| 272 |
+
"grad_norm": 2.972999095916748,
|
| 273 |
+
"learning_rate": 5.5791044776119405e-05,
|
| 274 |
+
"loss": 0.1892,
|
| 275 |
+
"step": 14500
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.46,
|
| 279 |
+
"grad_norm": 2.5483310222625732,
|
| 280 |
+
"learning_rate": 5.081592039800995e-05,
|
| 281 |
+
"loss": 0.1845,
|
| 282 |
+
"step": 15000
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.71,
|
| 286 |
+
"grad_norm": 2.5494730472564697,
|
| 287 |
+
"learning_rate": 4.58407960199005e-05,
|
| 288 |
+
"loss": 0.1848,
|
| 289 |
+
"step": 15500
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 7.96,
|
| 293 |
+
"grad_norm": 4.215138912200928,
|
| 294 |
+
"learning_rate": 4.086567164179105e-05,
|
| 295 |
+
"loss": 0.1812,
|
| 296 |
+
"step": 16000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 8.0,
|
| 300 |
+
"eval_accuracy": 0.9503206032229555,
|
| 301 |
+
"eval_loss": 0.1412646621465683,
|
| 302 |
+
"eval_runtime": 211.1854,
|
| 303 |
+
"eval_samples_per_second": 251.821,
|
| 304 |
+
"eval_steps_per_second": 31.479,
|
| 305 |
+
"step": 16080
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 8.21,
|
| 309 |
+
"grad_norm": 3.0078041553497314,
|
| 310 |
+
"learning_rate": 3.5900497512437814e-05,
|
| 311 |
+
"loss": 0.1719,
|
| 312 |
+
"step": 16500
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 8.46,
|
| 316 |
+
"grad_norm": 3.556603193283081,
|
| 317 |
+
"learning_rate": 3.092537313432836e-05,
|
| 318 |
+
"loss": 0.1717,
|
| 319 |
+
"step": 17000
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 8.71,
|
| 323 |
+
"grad_norm": 2.1547062397003174,
|
| 324 |
+
"learning_rate": 2.595024875621891e-05,
|
| 325 |
+
"loss": 0.1682,
|
| 326 |
+
"step": 17500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 8.96,
|
| 330 |
+
"grad_norm": 2.7211575508117676,
|
| 331 |
+
"learning_rate": 2.0975124378109454e-05,
|
| 332 |
+
"loss": 0.1632,
|
| 333 |
+
"step": 18000
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 9.0,
|
| 337 |
+
"eval_accuracy": 0.955303585867133,
|
| 338 |
+
"eval_loss": 0.13010455667972565,
|
| 339 |
+
"eval_runtime": 201.7962,
|
| 340 |
+
"eval_samples_per_second": 263.538,
|
| 341 |
+
"eval_steps_per_second": 32.944,
|
| 342 |
+
"step": 18090
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 9.2,
|
| 346 |
+
"grad_norm": 2.98066782951355,
|
| 347 |
+
"learning_rate": 1.600995024875622e-05,
|
| 348 |
+
"loss": 0.1626,
|
| 349 |
+
"step": 18500
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 9.45,
|
| 353 |
+
"grad_norm": 2.36365008354187,
|
| 354 |
+
"learning_rate": 1.1034825870646767e-05,
|
| 355 |
+
"loss": 0.1599,
|
| 356 |
+
"step": 19000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 9.7,
|
| 360 |
+
"grad_norm": 3.5116114616394043,
|
| 361 |
+
"learning_rate": 6.059701492537314e-06,
|
| 362 |
+
"loss": 0.1568,
|
| 363 |
+
"step": 19500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 9.95,
|
| 367 |
+
"grad_norm": 2.070329427719116,
|
| 368 |
+
"learning_rate": 1.0845771144278609e-06,
|
| 369 |
+
"loss": 0.1576,
|
| 370 |
+
"step": 20000
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 10.0,
|
| 374 |
+
"eval_accuracy": 0.9568830973467968,
|
| 375 |
+
"eval_loss": 0.12666618824005127,
|
| 376 |
+
"eval_runtime": 201.0618,
|
| 377 |
+
"eval_samples_per_second": 264.501,
|
| 378 |
+
"eval_steps_per_second": 33.064,
|
| 379 |
+
"step": 20100
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 10.0,
|
| 383 |
+
"step": 20100,
|
| 384 |
+
"total_flos": 7.572439992003492e+19,
|
| 385 |
+
"train_loss": 0.2540779383265557,
|
| 386 |
+
"train_runtime": 33188.7721,
|
| 387 |
+
"train_samples_per_second": 90.801,
|
| 388 |
+
"train_steps_per_second": 0.606
|
| 389 |
+
}
|
| 390 |
+
],
|
| 391 |
+
"logging_steps": 500,
|
| 392 |
+
"max_steps": 20100,
|
| 393 |
+
"num_input_tokens_seen": 0,
|
| 394 |
+
"num_train_epochs": 10,
|
| 395 |
+
"save_steps": 500,
|
| 396 |
+
"total_flos": 7.572439992003492e+19,
|
| 397 |
+
"train_batch_size": 150,
|
| 398 |
+
"trial_name": null,
|
| 399 |
+
"trial_params": null
|
| 400 |
+
}
|