End of training
Browse files- README.md +12 -46
- all_results.json +5 -5
- config.json +10 -10
- model.safetensors +2 -2
- runs/Mar25_09-23-36_plan.cs.vt.edu/events.out.tfevents.1711383837.plan.cs.vt.edu.376416.0 +3 -0
- runs/Mar25_09-32-34_plan.cs.vt.edu/events.out.tfevents.1711384364.plan.cs.vt.edu.376416.1 +3 -0
- runs/Mar25_09-32-34_plan.cs.vt.edu/events.out.tfevents.1711384513.plan.cs.vt.edu.376416.2 +3 -0
- runs/Mar25_09-32-34_plan.cs.vt.edu/events.out.tfevents.1711384569.plan.cs.vt.edu.376416.3 +3 -0
- runs/Mar25_09-32-34_plan.cs.vt.edu/events.out.tfevents.1711384600.plan.cs.vt.edu.376416.4 +3 -0
- runs/Mar25_09-32-34_plan.cs.vt.edu/events.out.tfevents.1711384646.plan.cs.vt.edu.376416.5 +3 -0
- runs/Mar25_09-40-56_plan.cs.vt.edu/events.out.tfevents.1711384864.plan.cs.vt.edu.394257.0 +3 -0
- runs/Mar25_09-40-56_plan.cs.vt.edu/events.out.tfevents.1711385275.plan.cs.vt.edu.394257.1 +3 -0
- test_results.json +5 -5
- trainer_state.json +106 -652
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
---
|
| 2 |
license: cc-by-nc-4.0
|
| 3 |
-
base_model: MCG-NJU/videomae-
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
|
@@ -15,9 +15,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# videomae-huge
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [MCG-NJU/videomae-
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss:
|
| 21 |
- Accuracy: 0.4286
|
| 22 |
|
| 23 |
## Model description
|
|
@@ -38,58 +38,24 @@ More information needed
|
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
- learning_rate: 5e-05
|
| 41 |
-
- train_batch_size:
|
| 42 |
-
- eval_batch_size:
|
| 43 |
- seed: 42
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: linear
|
| 46 |
- lr_scheduler_warmup_ratio: 0.1
|
| 47 |
-
- training_steps:
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 52 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 53 |
-
| 2.
|
| 54 |
-
|
|
| 55 |
-
|
|
| 56 |
-
|
|
| 57 |
-
|
|
| 58 |
-
|
|
| 59 |
-
| 1.726 | 6.03 | 49 | 2.2536 | 0.3214 |
|
| 60 |
-
| 1.3251 | 7.03 | 56 | 2.1674 | 0.3571 |
|
| 61 |
-
| 1.174 | 8.03 | 63 | 2.1128 | 0.3571 |
|
| 62 |
-
| 0.9108 | 9.03 | 70 | 2.0246 | 0.4286 |
|
| 63 |
-
| 0.7467 | 10.03 | 77 | 1.9720 | 0.3571 |
|
| 64 |
-
| 0.6224 | 11.03 | 84 | 1.9130 | 0.3929 |
|
| 65 |
-
| 0.4737 | 12.03 | 91 | 1.9136 | 0.3929 |
|
| 66 |
-
| 0.3682 | 13.03 | 98 | 1.8184 | 0.4643 |
|
| 67 |
-
| 0.2657 | 14.03 | 105 | 1.8734 | 0.3571 |
|
| 68 |
-
| 0.2085 | 15.03 | 112 | 1.8544 | 0.4286 |
|
| 69 |
-
| 0.175 | 16.03 | 119 | 1.8411 | 0.3929 |
|
| 70 |
-
| 0.1327 | 17.03 | 126 | 1.7585 | 0.3929 |
|
| 71 |
-
| 0.1001 | 18.03 | 133 | 1.8193 | 0.3929 |
|
| 72 |
-
| 0.0832 | 19.03 | 140 | 1.7595 | 0.3929 |
|
| 73 |
-
| 0.085 | 20.03 | 147 | 1.7836 | 0.4286 |
|
| 74 |
-
| 0.0614 | 21.03 | 154 | 1.7182 | 0.4286 |
|
| 75 |
-
| 0.0496 | 22.03 | 161 | 1.7845 | 0.4286 |
|
| 76 |
-
| 0.0474 | 23.03 | 168 | 1.7712 | 0.4643 |
|
| 77 |
-
| 0.0403 | 24.03 | 175 | 1.7409 | 0.4643 |
|
| 78 |
-
| 0.0395 | 25.03 | 182 | 1.7425 | 0.4643 |
|
| 79 |
-
| 0.0369 | 26.03 | 189 | 1.7532 | 0.4643 |
|
| 80 |
-
| 0.0346 | 27.03 | 196 | 1.7388 | 0.5 |
|
| 81 |
-
| 0.0321 | 28.03 | 203 | 1.7391 | 0.5 |
|
| 82 |
-
| 0.0314 | 29.03 | 210 | 1.7470 | 0.5 |
|
| 83 |
-
| 0.0313 | 30.03 | 217 | 1.7349 | 0.4643 |
|
| 84 |
-
| 0.0307 | 31.03 | 224 | 1.7574 | 0.4643 |
|
| 85 |
-
| 0.0283 | 32.03 | 231 | 1.7857 | 0.4286 |
|
| 86 |
-
| 0.0276 | 33.03 | 238 | 1.7865 | 0.4643 |
|
| 87 |
-
| 0.0257 | 34.03 | 245 | 1.7707 | 0.4643 |
|
| 88 |
-
| 0.0264 | 35.03 | 252 | 1.7683 | 0.4643 |
|
| 89 |
-
| 0.0254 | 36.03 | 259 | 1.7712 | 0.4643 |
|
| 90 |
-
| 0.0257 | 37.03 | 266 | 1.7736 | 0.4643 |
|
| 91 |
-
| 0.0246 | 38.03 | 273 | 1.7748 | 0.4643 |
|
| 92 |
-
| 0.0242 | 39.01 | 275 | 1.7750 | 0.4643 |
|
| 93 |
|
| 94 |
|
| 95 |
### Framework versions
|
|
|
|
| 1 |
---
|
| 2 |
license: cc-by-nc-4.0
|
| 3 |
+
base_model: MCG-NJU/videomae-base-finetuned-kinetics
|
| 4 |
tags:
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
|
|
|
| 15 |
|
| 16 |
# videomae-huge
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [MCG-NJU/videomae-base-finetuned-kinetics](https://huggingface.co/MCG-NJU/videomae-base-finetuned-kinetics) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 2.0714
|
| 21 |
- Accuracy: 0.4286
|
| 22 |
|
| 23 |
## Model description
|
|
|
|
| 38 |
|
| 39 |
The following hyperparameters were used during training:
|
| 40 |
- learning_rate: 5e-05
|
| 41 |
+
- train_batch_size: 8
|
| 42 |
+
- eval_batch_size: 8
|
| 43 |
- seed: 42
|
| 44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
- lr_scheduler_type: linear
|
| 46 |
- lr_scheduler_warmup_ratio: 0.1
|
| 47 |
+
- training_steps: 78
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 52 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 53 |
+
| 2.6392 | 0.18 | 14 | 2.5780 | 0.25 |
|
| 54 |
+
| 1.994 | 1.18 | 28 | 2.3995 | 0.3929 |
|
| 55 |
+
| 1.6374 | 2.18 | 42 | 2.3010 | 0.3929 |
|
| 56 |
+
| 1.124 | 3.18 | 56 | 2.2242 | 0.3929 |
|
| 57 |
+
| 0.9569 | 4.18 | 70 | 2.1825 | 0.4286 |
|
| 58 |
+
| 0.8862 | 5.1 | 78 | 2.1734 | 0.4286 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
|
| 60 |
|
| 61 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"eval_accuracy": 0.42857142857142855,
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 3.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.1,
|
| 3 |
"eval_accuracy": 0.42857142857142855,
|
| 4 |
+
"eval_loss": 2.0714259147644043,
|
| 5 |
+
"eval_runtime": 3.8707,
|
| 6 |
+
"eval_samples_per_second": 3.617,
|
| 7 |
+
"eval_steps_per_second": 0.517
|
| 8 |
}
|
config.json
CHANGED
|
@@ -1,16 +1,16 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "MCG-NJU/videomae-
|
| 3 |
"architectures": [
|
| 4 |
"VideoMAEForVideoClassification"
|
| 5 |
],
|
| 6 |
"attention_probs_dropout_prob": 0.0,
|
| 7 |
-
"decoder_hidden_size":
|
| 8 |
-
"decoder_intermediate_size":
|
| 9 |
-
"decoder_num_attention_heads":
|
| 10 |
-
"decoder_num_hidden_layers":
|
| 11 |
"hidden_act": "gelu",
|
| 12 |
"hidden_dropout_prob": 0.0,
|
| 13 |
-
"hidden_size":
|
| 14 |
"id2label": {
|
| 15 |
"0": "climb",
|
| 16 |
"1": "crawl",
|
|
@@ -29,7 +29,7 @@
|
|
| 29 |
},
|
| 30 |
"image_size": 224,
|
| 31 |
"initializer_range": 0.02,
|
| 32 |
-
"intermediate_size":
|
| 33 |
"label2id": {
|
| 34 |
"climb": 0,
|
| 35 |
"crawl": 1,
|
|
@@ -48,11 +48,11 @@
|
|
| 48 |
},
|
| 49 |
"layer_norm_eps": 1e-12,
|
| 50 |
"model_type": "videomae",
|
| 51 |
-
"norm_pix_loss":
|
| 52 |
-
"num_attention_heads":
|
| 53 |
"num_channels": 3,
|
| 54 |
"num_frames": 16,
|
| 55 |
-
"num_hidden_layers":
|
| 56 |
"patch_size": 16,
|
| 57 |
"problem_type": "single_label_classification",
|
| 58 |
"qkv_bias": true,
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "MCG-NJU/videomae-base-finetuned-kinetics",
|
| 3 |
"architectures": [
|
| 4 |
"VideoMAEForVideoClassification"
|
| 5 |
],
|
| 6 |
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"decoder_hidden_size": 384,
|
| 8 |
+
"decoder_intermediate_size": 1536,
|
| 9 |
+
"decoder_num_attention_heads": 6,
|
| 10 |
+
"decoder_num_hidden_layers": 4,
|
| 11 |
"hidden_act": "gelu",
|
| 12 |
"hidden_dropout_prob": 0.0,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
"id2label": {
|
| 15 |
"0": "climb",
|
| 16 |
"1": "crawl",
|
|
|
|
| 29 |
},
|
| 30 |
"image_size": 224,
|
| 31 |
"initializer_range": 0.02,
|
| 32 |
+
"intermediate_size": 3072,
|
| 33 |
"label2id": {
|
| 34 |
"climb": 0,
|
| 35 |
"crawl": 1,
|
|
|
|
| 48 |
},
|
| 49 |
"layer_norm_eps": 1e-12,
|
| 50 |
"model_type": "videomae",
|
| 51 |
+
"norm_pix_loss": false,
|
| 52 |
+
"num_attention_heads": 12,
|
| 53 |
"num_channels": 3,
|
| 54 |
"num_frames": 16,
|
| 55 |
+
"num_hidden_layers": 12,
|
| 56 |
"patch_size": 16,
|
| 57 |
"problem_type": "single_label_classification",
|
| 58 |
"qkv_bias": true,
|
model.safetensors
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 sha256:2123bd697e01c062399aa24d6818dc3064782f219a69cd69864c69d5aa3df57f
|
| 3 |
+
size 344974248
|
runs/Mar25_09-23-36_plan.cs.vt.edu/events.out.tfevents.1711383837.plan.cs.vt.edu.376416.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ef33a4dbd743b6b84fdcd8cdb9e0a35f7b911a4b769f96e2df816bbf6d90563
|
| 3 |
+
size 5074
|
runs/Mar25_09-32-34_plan.cs.vt.edu/events.out.tfevents.1711384364.plan.cs.vt.edu.376416.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:48239d8c28d82c026743bf94e14c63195d7cb740c2ad82d331d6143c62a2da95
|
| 3 |
+
size 5074
|
runs/Mar25_09-32-34_plan.cs.vt.edu/events.out.tfevents.1711384513.plan.cs.vt.edu.376416.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4df2ce2566dea9be0b2e6860e90bdc1a33374386cb436919b963fcbf95a87d43
|
| 3 |
+
size 5074
|
runs/Mar25_09-32-34_plan.cs.vt.edu/events.out.tfevents.1711384569.plan.cs.vt.edu.376416.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:73c885e9631091493a0606fff30004e065409c2e1353d9ed405f741617bb4f26
|
| 3 |
+
size 5123
|
runs/Mar25_09-32-34_plan.cs.vt.edu/events.out.tfevents.1711384600.plan.cs.vt.edu.376416.4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64e24efa2ca59f331be4c14c47350f77999cfbe4f00c608a823142bbdf3ce2f2
|
| 3 |
+
size 5123
|
runs/Mar25_09-32-34_plan.cs.vt.edu/events.out.tfevents.1711384646.plan.cs.vt.edu.376416.5
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3e720408dfcd55af363cf818d1acb9244f9a0bdb152ec847b1d4add5b542f54
|
| 3 |
+
size 5123
|
runs/Mar25_09-40-56_plan.cs.vt.edu/events.out.tfevents.1711384864.plan.cs.vt.edu.394257.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:992c87eb2cb0ee976f99118e93fa327020001d7921fc68c3c8a0cb00b341f932
|
| 3 |
+
size 9634
|
runs/Mar25_09-40-56_plan.cs.vt.edu/events.out.tfevents.1711385275.plan.cs.vt.edu.394257.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:26face0e8f8dc66b68723685ae10279d5420f09069f04872a3b8bd1973ce6687
|
| 3 |
+
size 722
|
test_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"eval_accuracy": 0.42857142857142855,
|
| 4 |
-
"eval_loss":
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 3.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 5.1,
|
| 3 |
"eval_accuracy": 0.42857142857142855,
|
| 4 |
+
"eval_loss": 2.0714259147644043,
|
| 5 |
+
"eval_runtime": 3.8707,
|
| 6 |
+
"eval_samples_per_second": 3.617,
|
| 7 |
+
"eval_steps_per_second": 0.517
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,738 +1,192 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "Tianjiao-Yu/videomae-huge/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 |
-
"learning_rate":
|
| 14 |
-
"loss": 2.
|
| 15 |
"step": 5
|
| 16 |
},
|
| 17 |
{
|
| 18 |
-
"epoch": 0.
|
| 19 |
-
"
|
| 20 |
-
"
|
| 21 |
-
"eval_runtime": 8.9709,
|
| 22 |
-
"eval_samples_per_second": 3.121,
|
| 23 |
-
"eval_steps_per_second": 0.223,
|
| 24 |
-
"step": 7
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 1.01,
|
| 28 |
-
"learning_rate": 1.785714285714286e-05,
|
| 29 |
-
"loss": 2.6262,
|
| 30 |
"step": 10
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch":
|
| 34 |
-
"eval_accuracy": 0.
|
| 35 |
-
"eval_loss": 2.
|
| 36 |
-
"eval_runtime": 7.
|
| 37 |
-
"eval_samples_per_second": 3.
|
| 38 |
-
"eval_steps_per_second": 0.
|
| 39 |
"step": 14
|
| 40 |
},
|
| 41 |
{
|
| 42 |
-
"epoch":
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 2.
|
| 45 |
"step": 15
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 2.
|
| 51 |
"step": 20
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch":
|
| 55 |
-
"
|
| 56 |
-
"
|
| 57 |
-
"eval_runtime": 8.1508,
|
| 58 |
-
"eval_samples_per_second": 3.435,
|
| 59 |
-
"eval_steps_per_second": 0.245,
|
| 60 |
-
"step": 21
|
| 61 |
-
},
|
| 62 |
-
{
|
| 63 |
-
"epoch": 3.01,
|
| 64 |
-
"learning_rate": 4.464285714285715e-05,
|
| 65 |
-
"loss": 2.4135,
|
| 66 |
"step": 25
|
| 67 |
},
|
| 68 |
{
|
| 69 |
-
"epoch":
|
| 70 |
-
"eval_accuracy": 0.
|
| 71 |
-
"eval_loss": 2.
|
| 72 |
-
"eval_runtime": 7.
|
| 73 |
-
"eval_samples_per_second": 3.
|
| 74 |
-
"eval_steps_per_second": 0.
|
| 75 |
"step": 28
|
| 76 |
},
|
| 77 |
{
|
| 78 |
-
"epoch":
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss":
|
| 81 |
"step": 30
|
| 82 |
},
|
| 83 |
{
|
| 84 |
-
"epoch":
|
| 85 |
-
"learning_rate":
|
| 86 |
-
"loss":
|
| 87 |
"step": 35
|
| 88 |
},
|
| 89 |
{
|
| 90 |
-
"epoch":
|
| 91 |
-
"
|
| 92 |
-
"
|
| 93 |
-
"eval_runtime": 7.6554,
|
| 94 |
-
"eval_samples_per_second": 3.658,
|
| 95 |
-
"eval_steps_per_second": 0.261,
|
| 96 |
-
"step": 35
|
| 97 |
-
},
|
| 98 |
-
{
|
| 99 |
-
"epoch": 5.02,
|
| 100 |
-
"learning_rate": 4.757085020242915e-05,
|
| 101 |
-
"loss": 1.848,
|
| 102 |
"step": 40
|
| 103 |
},
|
| 104 |
{
|
| 105 |
-
"epoch":
|
| 106 |
-
"eval_accuracy": 0.
|
| 107 |
-
"eval_loss": 2.
|
| 108 |
-
"eval_runtime": 7.
|
| 109 |
-
"eval_samples_per_second": 3.
|
| 110 |
-
"eval_steps_per_second": 0.
|
| 111 |
"step": 42
|
| 112 |
},
|
| 113 |
{
|
| 114 |
-
"epoch":
|
| 115 |
-
"learning_rate":
|
| 116 |
-
"loss": 1.
|
| 117 |
"step": 45
|
| 118 |
},
|
| 119 |
{
|
| 120 |
-
"epoch":
|
| 121 |
-
"
|
| 122 |
-
"
|
| 123 |
-
"eval_runtime": 7.0278,
|
| 124 |
-
"eval_samples_per_second": 3.984,
|
| 125 |
-
"eval_steps_per_second": 0.285,
|
| 126 |
-
"step": 49
|
| 127 |
-
},
|
| 128 |
-
{
|
| 129 |
-
"epoch": 7.0,
|
| 130 |
-
"learning_rate": 4.5546558704453443e-05,
|
| 131 |
-
"loss": 1.5657,
|
| 132 |
"step": 50
|
| 133 |
},
|
| 134 |
{
|
| 135 |
-
"epoch":
|
| 136 |
-
"learning_rate":
|
| 137 |
-
"loss": 1.
|
| 138 |
"step": 55
|
| 139 |
},
|
| 140 |
{
|
| 141 |
-
"epoch":
|
| 142 |
-
"eval_accuracy": 0.
|
| 143 |
-
"eval_loss": 2.
|
| 144 |
-
"eval_runtime":
|
| 145 |
-
"eval_samples_per_second": 3.
|
| 146 |
-
"eval_steps_per_second": 0.
|
| 147 |
"step": 56
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"epoch":
|
| 151 |
-
"learning_rate":
|
| 152 |
-
"loss": 1.
|
| 153 |
"step": 60
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"epoch":
|
| 157 |
-
"
|
| 158 |
-
"
|
| 159 |
-
"eval_runtime": 8.0792,
|
| 160 |
-
"eval_samples_per_second": 3.466,
|
| 161 |
-
"eval_steps_per_second": 0.248,
|
| 162 |
-
"step": 63
|
| 163 |
-
},
|
| 164 |
-
{
|
| 165 |
-
"epoch": 9.01,
|
| 166 |
-
"learning_rate": 4.251012145748988e-05,
|
| 167 |
-
"loss": 0.9546,
|
| 168 |
"step": 65
|
| 169 |
},
|
| 170 |
{
|
| 171 |
-
"epoch":
|
| 172 |
-
"learning_rate":
|
| 173 |
-
"loss": 0.
|
| 174 |
"step": 70
|
| 175 |
},
|
| 176 |
{
|
| 177 |
-
"epoch":
|
| 178 |
"eval_accuracy": 0.42857142857142855,
|
| 179 |
-
"eval_loss": 2.
|
| 180 |
-
"eval_runtime": 7.
|
| 181 |
-
"eval_samples_per_second": 3.
|
| 182 |
-
"eval_steps_per_second": 0.
|
| 183 |
"step": 70
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"epoch":
|
| 187 |
-
"learning_rate":
|
| 188 |
-
"loss": 0.
|
| 189 |
"step": 75
|
| 190 |
},
|
| 191 |
{
|
| 192 |
-
"epoch":
|
| 193 |
-
"eval_accuracy": 0.35714285714285715,
|
| 194 |
-
"eval_loss": 1.9719778299331665,
|
| 195 |
-
"eval_runtime": 7.6011,
|
| 196 |
-
"eval_samples_per_second": 3.684,
|
| 197 |
-
"eval_steps_per_second": 0.263,
|
| 198 |
-
"step": 77
|
| 199 |
-
},
|
| 200 |
-
{
|
| 201 |
-
"epoch": 11.01,
|
| 202 |
-
"learning_rate": 3.9473684210526316e-05,
|
| 203 |
-
"loss": 0.6224,
|
| 204 |
-
"step": 80
|
| 205 |
-
},
|
| 206 |
-
{
|
| 207 |
-
"epoch": 11.03,
|
| 208 |
-
"eval_accuracy": 0.39285714285714285,
|
| 209 |
-
"eval_loss": 1.9129974842071533,
|
| 210 |
-
"eval_runtime": 7.7956,
|
| 211 |
-
"eval_samples_per_second": 3.592,
|
| 212 |
-
"eval_steps_per_second": 0.257,
|
| 213 |
-
"step": 84
|
| 214 |
-
},
|
| 215 |
-
{
|
| 216 |
-
"epoch": 12.0,
|
| 217 |
-
"learning_rate": 3.846153846153846e-05,
|
| 218 |
-
"loss": 0.5523,
|
| 219 |
-
"step": 85
|
| 220 |
-
},
|
| 221 |
-
{
|
| 222 |
-
"epoch": 12.02,
|
| 223 |
-
"learning_rate": 3.744939271255061e-05,
|
| 224 |
-
"loss": 0.4737,
|
| 225 |
-
"step": 90
|
| 226 |
-
},
|
| 227 |
-
{
|
| 228 |
-
"epoch": 12.03,
|
| 229 |
-
"eval_accuracy": 0.39285714285714285,
|
| 230 |
-
"eval_loss": 1.9135581254959106,
|
| 231 |
-
"eval_runtime": 8.3212,
|
| 232 |
-
"eval_samples_per_second": 3.365,
|
| 233 |
-
"eval_steps_per_second": 0.24,
|
| 234 |
-
"step": 91
|
| 235 |
-
},
|
| 236 |
-
{
|
| 237 |
-
"epoch": 13.01,
|
| 238 |
-
"learning_rate": 3.6437246963562756e-05,
|
| 239 |
-
"loss": 0.3682,
|
| 240 |
-
"step": 95
|
| 241 |
-
},
|
| 242 |
-
{
|
| 243 |
-
"epoch": 13.03,
|
| 244 |
-
"eval_accuracy": 0.4642857142857143,
|
| 245 |
-
"eval_loss": 1.8183634281158447,
|
| 246 |
-
"eval_runtime": 8.3757,
|
| 247 |
-
"eval_samples_per_second": 3.343,
|
| 248 |
-
"eval_steps_per_second": 0.239,
|
| 249 |
-
"step": 98
|
| 250 |
-
},
|
| 251 |
-
{
|
| 252 |
-
"epoch": 14.01,
|
| 253 |
-
"learning_rate": 3.54251012145749e-05,
|
| 254 |
-
"loss": 0.3151,
|
| 255 |
-
"step": 100
|
| 256 |
-
},
|
| 257 |
-
{
|
| 258 |
-
"epoch": 14.03,
|
| 259 |
-
"learning_rate": 3.441295546558704e-05,
|
| 260 |
-
"loss": 0.2657,
|
| 261 |
-
"step": 105
|
| 262 |
-
},
|
| 263 |
-
{
|
| 264 |
-
"epoch": 14.03,
|
| 265 |
-
"eval_accuracy": 0.35714285714285715,
|
| 266 |
-
"eval_loss": 1.8734323978424072,
|
| 267 |
-
"eval_runtime": 8.0062,
|
| 268 |
-
"eval_samples_per_second": 3.497,
|
| 269 |
-
"eval_steps_per_second": 0.25,
|
| 270 |
-
"step": 105
|
| 271 |
-
},
|
| 272 |
-
{
|
| 273 |
-
"epoch": 15.02,
|
| 274 |
-
"learning_rate": 3.340080971659919e-05,
|
| 275 |
-
"loss": 0.2085,
|
| 276 |
-
"step": 110
|
| 277 |
-
},
|
| 278 |
-
{
|
| 279 |
-
"epoch": 15.03,
|
| 280 |
-
"eval_accuracy": 0.42857142857142855,
|
| 281 |
-
"eval_loss": 1.8544093370437622,
|
| 282 |
-
"eval_runtime": 7.1501,
|
| 283 |
-
"eval_samples_per_second": 3.916,
|
| 284 |
-
"eval_steps_per_second": 0.28,
|
| 285 |
-
"step": 112
|
| 286 |
-
},
|
| 287 |
-
{
|
| 288 |
-
"epoch": 16.01,
|
| 289 |
-
"learning_rate": 3.2388663967611336e-05,
|
| 290 |
-
"loss": 0.175,
|
| 291 |
-
"step": 115
|
| 292 |
-
},
|
| 293 |
-
{
|
| 294 |
-
"epoch": 16.03,
|
| 295 |
-
"eval_accuracy": 0.39285714285714285,
|
| 296 |
-
"eval_loss": 1.8410834074020386,
|
| 297 |
-
"eval_runtime": 7.4699,
|
| 298 |
-
"eval_samples_per_second": 3.748,
|
| 299 |
-
"eval_steps_per_second": 0.268,
|
| 300 |
-
"step": 119
|
| 301 |
-
},
|
| 302 |
-
{
|
| 303 |
-
"epoch": 17.0,
|
| 304 |
-
"learning_rate": 3.137651821862348e-05,
|
| 305 |
-
"loss": 0.1416,
|
| 306 |
-
"step": 120
|
| 307 |
-
},
|
| 308 |
-
{
|
| 309 |
-
"epoch": 17.02,
|
| 310 |
-
"learning_rate": 3.0364372469635626e-05,
|
| 311 |
-
"loss": 0.1327,
|
| 312 |
-
"step": 125
|
| 313 |
-
},
|
| 314 |
-
{
|
| 315 |
-
"epoch": 17.03,
|
| 316 |
-
"eval_accuracy": 0.39285714285714285,
|
| 317 |
-
"eval_loss": 1.7585405111312866,
|
| 318 |
-
"eval_runtime": 7.2596,
|
| 319 |
-
"eval_samples_per_second": 3.857,
|
| 320 |
-
"eval_steps_per_second": 0.275,
|
| 321 |
-
"step": 126
|
| 322 |
-
},
|
| 323 |
-
{
|
| 324 |
-
"epoch": 18.01,
|
| 325 |
-
"learning_rate": 2.9352226720647776e-05,
|
| 326 |
-
"loss": 0.1001,
|
| 327 |
-
"step": 130
|
| 328 |
-
},
|
| 329 |
-
{
|
| 330 |
-
"epoch": 18.03,
|
| 331 |
-
"eval_accuracy": 0.39285714285714285,
|
| 332 |
-
"eval_loss": 1.8193204402923584,
|
| 333 |
-
"eval_runtime": 7.6478,
|
| 334 |
-
"eval_samples_per_second": 3.661,
|
| 335 |
-
"eval_steps_per_second": 0.262,
|
| 336 |
-
"step": 133
|
| 337 |
-
},
|
| 338 |
-
{
|
| 339 |
-
"epoch": 19.01,
|
| 340 |
-
"learning_rate": 2.8340080971659922e-05,
|
| 341 |
-
"loss": 0.094,
|
| 342 |
-
"step": 135
|
| 343 |
-
},
|
| 344 |
-
{
|
| 345 |
-
"epoch": 19.03,
|
| 346 |
-
"learning_rate": 2.732793522267207e-05,
|
| 347 |
-
"loss": 0.0832,
|
| 348 |
-
"step": 140
|
| 349 |
-
},
|
| 350 |
-
{
|
| 351 |
-
"epoch": 19.03,
|
| 352 |
-
"eval_accuracy": 0.39285714285714285,
|
| 353 |
-
"eval_loss": 1.759466528892517,
|
| 354 |
-
"eval_runtime": 7.1794,
|
| 355 |
-
"eval_samples_per_second": 3.9,
|
| 356 |
-
"eval_steps_per_second": 0.279,
|
| 357 |
-
"step": 140
|
| 358 |
-
},
|
| 359 |
-
{
|
| 360 |
-
"epoch": 20.02,
|
| 361 |
-
"learning_rate": 2.6315789473684212e-05,
|
| 362 |
-
"loss": 0.085,
|
| 363 |
-
"step": 145
|
| 364 |
-
},
|
| 365 |
-
{
|
| 366 |
-
"epoch": 20.03,
|
| 367 |
-
"eval_accuracy": 0.42857142857142855,
|
| 368 |
-
"eval_loss": 1.7836247682571411,
|
| 369 |
-
"eval_runtime": 7.7103,
|
| 370 |
-
"eval_samples_per_second": 3.631,
|
| 371 |
-
"eval_steps_per_second": 0.259,
|
| 372 |
-
"step": 147
|
| 373 |
-
},
|
| 374 |
-
{
|
| 375 |
-
"epoch": 21.01,
|
| 376 |
-
"learning_rate": 2.530364372469636e-05,
|
| 377 |
-
"loss": 0.0614,
|
| 378 |
-
"step": 150
|
| 379 |
-
},
|
| 380 |
-
{
|
| 381 |
-
"epoch": 21.03,
|
| 382 |
-
"eval_accuracy": 0.42857142857142855,
|
| 383 |
-
"eval_loss": 1.7181528806686401,
|
| 384 |
-
"eval_runtime": 7.4571,
|
| 385 |
-
"eval_samples_per_second": 3.755,
|
| 386 |
-
"eval_steps_per_second": 0.268,
|
| 387 |
-
"step": 154
|
| 388 |
-
},
|
| 389 |
-
{
|
| 390 |
-
"epoch": 22.0,
|
| 391 |
-
"learning_rate": 2.4291497975708502e-05,
|
| 392 |
-
"loss": 0.0604,
|
| 393 |
-
"step": 155
|
| 394 |
-
},
|
| 395 |
-
{
|
| 396 |
-
"epoch": 22.02,
|
| 397 |
-
"learning_rate": 2.327935222672065e-05,
|
| 398 |
-
"loss": 0.0496,
|
| 399 |
-
"step": 160
|
| 400 |
-
},
|
| 401 |
-
{
|
| 402 |
-
"epoch": 22.03,
|
| 403 |
-
"eval_accuracy": 0.42857142857142855,
|
| 404 |
-
"eval_loss": 1.7844511270523071,
|
| 405 |
-
"eval_runtime": 7.281,
|
| 406 |
-
"eval_samples_per_second": 3.846,
|
| 407 |
-
"eval_steps_per_second": 0.275,
|
| 408 |
-
"step": 161
|
| 409 |
-
},
|
| 410 |
-
{
|
| 411 |
-
"epoch": 23.01,
|
| 412 |
-
"learning_rate": 2.2267206477732795e-05,
|
| 413 |
-
"loss": 0.0474,
|
| 414 |
-
"step": 165
|
| 415 |
-
},
|
| 416 |
-
{
|
| 417 |
-
"epoch": 23.03,
|
| 418 |
-
"eval_accuracy": 0.4642857142857143,
|
| 419 |
-
"eval_loss": 1.771189570426941,
|
| 420 |
-
"eval_runtime": 7.0825,
|
| 421 |
-
"eval_samples_per_second": 3.953,
|
| 422 |
-
"eval_steps_per_second": 0.282,
|
| 423 |
-
"step": 168
|
| 424 |
-
},
|
| 425 |
-
{
|
| 426 |
-
"epoch": 24.01,
|
| 427 |
-
"learning_rate": 2.125506072874494e-05,
|
| 428 |
-
"loss": 0.0423,
|
| 429 |
-
"step": 170
|
| 430 |
-
},
|
| 431 |
-
{
|
| 432 |
-
"epoch": 24.03,
|
| 433 |
-
"learning_rate": 2.0242914979757085e-05,
|
| 434 |
-
"loss": 0.0403,
|
| 435 |
-
"step": 175
|
| 436 |
-
},
|
| 437 |
-
{
|
| 438 |
-
"epoch": 24.03,
|
| 439 |
-
"eval_accuracy": 0.4642857142857143,
|
| 440 |
-
"eval_loss": 1.740900993347168,
|
| 441 |
-
"eval_runtime": 6.8813,
|
| 442 |
-
"eval_samples_per_second": 4.069,
|
| 443 |
-
"eval_steps_per_second": 0.291,
|
| 444 |
-
"step": 175
|
| 445 |
-
},
|
| 446 |
-
{
|
| 447 |
-
"epoch": 25.02,
|
| 448 |
-
"learning_rate": 1.923076923076923e-05,
|
| 449 |
-
"loss": 0.0395,
|
| 450 |
-
"step": 180
|
| 451 |
-
},
|
| 452 |
-
{
|
| 453 |
-
"epoch": 25.03,
|
| 454 |
-
"eval_accuracy": 0.4642857142857143,
|
| 455 |
-
"eval_loss": 1.7425283193588257,
|
| 456 |
-
"eval_runtime": 7.2597,
|
| 457 |
-
"eval_samples_per_second": 3.857,
|
| 458 |
-
"eval_steps_per_second": 0.275,
|
| 459 |
-
"step": 182
|
| 460 |
-
},
|
| 461 |
-
{
|
| 462 |
-
"epoch": 26.01,
|
| 463 |
-
"learning_rate": 1.8218623481781378e-05,
|
| 464 |
-
"loss": 0.0369,
|
| 465 |
-
"step": 185
|
| 466 |
-
},
|
| 467 |
-
{
|
| 468 |
-
"epoch": 26.03,
|
| 469 |
-
"eval_accuracy": 0.4642857142857143,
|
| 470 |
-
"eval_loss": 1.7531570196151733,
|
| 471 |
-
"eval_runtime": 6.8458,
|
| 472 |
-
"eval_samples_per_second": 4.09,
|
| 473 |
-
"eval_steps_per_second": 0.292,
|
| 474 |
-
"step": 189
|
| 475 |
-
},
|
| 476 |
-
{
|
| 477 |
-
"epoch": 27.0,
|
| 478 |
-
"learning_rate": 1.720647773279352e-05,
|
| 479 |
-
"loss": 0.0358,
|
| 480 |
-
"step": 190
|
| 481 |
-
},
|
| 482 |
-
{
|
| 483 |
-
"epoch": 27.02,
|
| 484 |
-
"learning_rate": 1.6194331983805668e-05,
|
| 485 |
-
"loss": 0.0346,
|
| 486 |
-
"step": 195
|
| 487 |
-
},
|
| 488 |
-
{
|
| 489 |
-
"epoch": 27.03,
|
| 490 |
-
"eval_accuracy": 0.5,
|
| 491 |
-
"eval_loss": 1.7388352155685425,
|
| 492 |
-
"eval_runtime": 7.2714,
|
| 493 |
-
"eval_samples_per_second": 3.851,
|
| 494 |
-
"eval_steps_per_second": 0.275,
|
| 495 |
-
"step": 196
|
| 496 |
-
},
|
| 497 |
-
{
|
| 498 |
-
"epoch": 28.01,
|
| 499 |
-
"learning_rate": 1.5182186234817813e-05,
|
| 500 |
-
"loss": 0.0321,
|
| 501 |
-
"step": 200
|
| 502 |
-
},
|
| 503 |
-
{
|
| 504 |
-
"epoch": 28.03,
|
| 505 |
-
"eval_accuracy": 0.5,
|
| 506 |
-
"eval_loss": 1.7390631437301636,
|
| 507 |
-
"eval_runtime": 7.0464,
|
| 508 |
-
"eval_samples_per_second": 3.974,
|
| 509 |
-
"eval_steps_per_second": 0.284,
|
| 510 |
-
"step": 203
|
| 511 |
-
},
|
| 512 |
-
{
|
| 513 |
-
"epoch": 29.01,
|
| 514 |
-
"learning_rate": 1.4170040485829961e-05,
|
| 515 |
-
"loss": 0.0351,
|
| 516 |
-
"step": 205
|
| 517 |
-
},
|
| 518 |
-
{
|
| 519 |
-
"epoch": 29.03,
|
| 520 |
-
"learning_rate": 1.3157894736842106e-05,
|
| 521 |
-
"loss": 0.0314,
|
| 522 |
-
"step": 210
|
| 523 |
-
},
|
| 524 |
-
{
|
| 525 |
-
"epoch": 29.03,
|
| 526 |
-
"eval_accuracy": 0.5,
|
| 527 |
-
"eval_loss": 1.7469881772994995,
|
| 528 |
-
"eval_runtime": 6.6833,
|
| 529 |
-
"eval_samples_per_second": 4.19,
|
| 530 |
-
"eval_steps_per_second": 0.299,
|
| 531 |
-
"step": 210
|
| 532 |
-
},
|
| 533 |
-
{
|
| 534 |
-
"epoch": 30.02,
|
| 535 |
-
"learning_rate": 1.2145748987854251e-05,
|
| 536 |
-
"loss": 0.0313,
|
| 537 |
-
"step": 215
|
| 538 |
-
},
|
| 539 |
-
{
|
| 540 |
-
"epoch": 30.03,
|
| 541 |
-
"eval_accuracy": 0.4642857142857143,
|
| 542 |
-
"eval_loss": 1.7349082231521606,
|
| 543 |
-
"eval_runtime": 6.7759,
|
| 544 |
-
"eval_samples_per_second": 4.132,
|
| 545 |
-
"eval_steps_per_second": 0.295,
|
| 546 |
-
"step": 217
|
| 547 |
-
},
|
| 548 |
-
{
|
| 549 |
-
"epoch": 31.01,
|
| 550 |
-
"learning_rate": 1.1133603238866398e-05,
|
| 551 |
-
"loss": 0.0307,
|
| 552 |
-
"step": 220
|
| 553 |
-
},
|
| 554 |
-
{
|
| 555 |
-
"epoch": 31.03,
|
| 556 |
-
"eval_accuracy": 0.4642857142857143,
|
| 557 |
-
"eval_loss": 1.7574304342269897,
|
| 558 |
-
"eval_runtime": 6.501,
|
| 559 |
-
"eval_samples_per_second": 4.307,
|
| 560 |
-
"eval_steps_per_second": 0.308,
|
| 561 |
-
"step": 224
|
| 562 |
-
},
|
| 563 |
-
{
|
| 564 |
-
"epoch": 32.0,
|
| 565 |
-
"learning_rate": 1.0121457489878542e-05,
|
| 566 |
-
"loss": 0.0296,
|
| 567 |
-
"step": 225
|
| 568 |
-
},
|
| 569 |
-
{
|
| 570 |
-
"epoch": 32.02,
|
| 571 |
-
"learning_rate": 9.109311740890689e-06,
|
| 572 |
-
"loss": 0.0283,
|
| 573 |
-
"step": 230
|
| 574 |
-
},
|
| 575 |
-
{
|
| 576 |
-
"epoch": 32.03,
|
| 577 |
"eval_accuracy": 0.42857142857142855,
|
| 578 |
-
"eval_loss":
|
| 579 |
-
"eval_runtime":
|
| 580 |
-
"eval_samples_per_second":
|
| 581 |
-
"eval_steps_per_second": 0.
|
| 582 |
-
"step":
|
| 583 |
-
},
|
| 584 |
-
{
|
| 585 |
-
"epoch": 33.01,
|
| 586 |
-
"learning_rate": 8.097165991902834e-06,
|
| 587 |
-
"loss": 0.0276,
|
| 588 |
-
"step": 235
|
| 589 |
-
},
|
| 590 |
-
{
|
| 591 |
-
"epoch": 33.03,
|
| 592 |
-
"eval_accuracy": 0.4642857142857143,
|
| 593 |
-
"eval_loss": 1.786450982093811,
|
| 594 |
-
"eval_runtime": 7.4349,
|
| 595 |
-
"eval_samples_per_second": 3.766,
|
| 596 |
-
"eval_steps_per_second": 0.269,
|
| 597 |
-
"step": 238
|
| 598 |
-
},
|
| 599 |
-
{
|
| 600 |
-
"epoch": 34.01,
|
| 601 |
-
"learning_rate": 7.0850202429149805e-06,
|
| 602 |
-
"loss": 0.0282,
|
| 603 |
-
"step": 240
|
| 604 |
-
},
|
| 605 |
-
{
|
| 606 |
-
"epoch": 34.03,
|
| 607 |
-
"learning_rate": 6.0728744939271254e-06,
|
| 608 |
-
"loss": 0.0257,
|
| 609 |
-
"step": 245
|
| 610 |
-
},
|
| 611 |
-
{
|
| 612 |
-
"epoch": 34.03,
|
| 613 |
-
"eval_accuracy": 0.4642857142857143,
|
| 614 |
-
"eval_loss": 1.7707021236419678,
|
| 615 |
-
"eval_runtime": 7.9939,
|
| 616 |
-
"eval_samples_per_second": 3.503,
|
| 617 |
-
"eval_steps_per_second": 0.25,
|
| 618 |
-
"step": 245
|
| 619 |
-
},
|
| 620 |
-
{
|
| 621 |
-
"epoch": 35.02,
|
| 622 |
-
"learning_rate": 5.060728744939271e-06,
|
| 623 |
-
"loss": 0.0264,
|
| 624 |
-
"step": 250
|
| 625 |
},
|
| 626 |
{
|
| 627 |
-
"epoch":
|
| 628 |
-
"
|
| 629 |
-
"
|
| 630 |
-
"
|
| 631 |
-
"
|
| 632 |
-
"
|
| 633 |
-
"
|
| 634 |
},
|
| 635 |
{
|
| 636 |
-
"epoch":
|
| 637 |
-
"learning_rate": 4.048582995951417e-06,
|
| 638 |
-
"loss": 0.0254,
|
| 639 |
-
"step": 255
|
| 640 |
-
},
|
| 641 |
-
{
|
| 642 |
-
"epoch": 36.03,
|
| 643 |
-
"eval_accuracy": 0.4642857142857143,
|
| 644 |
-
"eval_loss": 1.7711747884750366,
|
| 645 |
-
"eval_runtime": 7.529,
|
| 646 |
-
"eval_samples_per_second": 3.719,
|
| 647 |
-
"eval_steps_per_second": 0.266,
|
| 648 |
-
"step": 259
|
| 649 |
-
},
|
| 650 |
-
{
|
| 651 |
-
"epoch": 37.0,
|
| 652 |
-
"learning_rate": 3.0364372469635627e-06,
|
| 653 |
-
"loss": 0.0247,
|
| 654 |
-
"step": 260
|
| 655 |
-
},
|
| 656 |
-
{
|
| 657 |
-
"epoch": 37.02,
|
| 658 |
-
"learning_rate": 2.0242914979757085e-06,
|
| 659 |
-
"loss": 0.0257,
|
| 660 |
-
"step": 265
|
| 661 |
-
},
|
| 662 |
-
{
|
| 663 |
-
"epoch": 37.03,
|
| 664 |
-
"eval_accuracy": 0.4642857142857143,
|
| 665 |
-
"eval_loss": 1.7735536098480225,
|
| 666 |
-
"eval_runtime": 7.5429,
|
| 667 |
-
"eval_samples_per_second": 3.712,
|
| 668 |
-
"eval_steps_per_second": 0.265,
|
| 669 |
-
"step": 266
|
| 670 |
-
},
|
| 671 |
-
{
|
| 672 |
-
"epoch": 38.01,
|
| 673 |
-
"learning_rate": 1.0121457489878542e-06,
|
| 674 |
-
"loss": 0.0246,
|
| 675 |
-
"step": 270
|
| 676 |
-
},
|
| 677 |
-
{
|
| 678 |
-
"epoch": 38.03,
|
| 679 |
-
"eval_accuracy": 0.4642857142857143,
|
| 680 |
-
"eval_loss": 1.7748419046401978,
|
| 681 |
-
"eval_runtime": 7.4873,
|
| 682 |
-
"eval_samples_per_second": 3.74,
|
| 683 |
-
"eval_steps_per_second": 0.267,
|
| 684 |
-
"step": 273
|
| 685 |
-
},
|
| 686 |
-
{
|
| 687 |
-
"epoch": 39.01,
|
| 688 |
-
"learning_rate": 0.0,
|
| 689 |
-
"loss": 0.0242,
|
| 690 |
-
"step": 275
|
| 691 |
-
},
|
| 692 |
-
{
|
| 693 |
-
"epoch": 39.01,
|
| 694 |
-
"eval_accuracy": 0.4642857142857143,
|
| 695 |
-
"eval_loss": 1.774975061416626,
|
| 696 |
-
"eval_runtime": 7.7693,
|
| 697 |
-
"eval_samples_per_second": 3.604,
|
| 698 |
-
"eval_steps_per_second": 0.257,
|
| 699 |
-
"step": 275
|
| 700 |
-
},
|
| 701 |
-
{
|
| 702 |
-
"epoch": 39.01,
|
| 703 |
-
"step": 275,
|
| 704 |
-
"total_flos": 3.980465870018229e+19,
|
| 705 |
-
"train_loss": 0.5794233712283048,
|
| 706 |
-
"train_runtime": 2037.2017,
|
| 707 |
-
"train_samples_per_second": 2.16,
|
| 708 |
-
"train_steps_per_second": 0.135
|
| 709 |
-
},
|
| 710 |
-
{
|
| 711 |
-
"epoch": 39.01,
|
| 712 |
"eval_accuracy": 0.42857142857142855,
|
| 713 |
-
"eval_loss":
|
| 714 |
-
"eval_runtime":
|
| 715 |
-
"eval_samples_per_second":
|
| 716 |
-
"eval_steps_per_second": 0.
|
| 717 |
-
"step":
|
| 718 |
},
|
| 719 |
{
|
| 720 |
-
"epoch":
|
| 721 |
"eval_accuracy": 0.42857142857142855,
|
| 722 |
-
"eval_loss":
|
| 723 |
-
"eval_runtime":
|
| 724 |
-
"eval_samples_per_second": 3.
|
| 725 |
-
"eval_steps_per_second": 0.
|
| 726 |
-
"step":
|
| 727 |
}
|
| 728 |
],
|
| 729 |
"logging_steps": 5,
|
| 730 |
-
"max_steps":
|
| 731 |
"num_input_tokens_seen": 0,
|
| 732 |
"num_train_epochs": 9223372036854775807,
|
| 733 |
"save_steps": 500,
|
| 734 |
-
"total_flos":
|
| 735 |
-
"train_batch_size":
|
| 736 |
"trial_name": null,
|
| 737 |
"trial_params": null
|
| 738 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.42857142857142855,
|
| 3 |
+
"best_model_checkpoint": "Tianjiao-Yu/videomae-huge/checkpoint-70",
|
| 4 |
+
"epoch": 5.102564102564102,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 78,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.06,
|
| 13 |
+
"learning_rate": 3.125e-05,
|
| 14 |
+
"loss": 2.66,
|
| 15 |
"step": 5
|
| 16 |
},
|
| 17 |
{
|
| 18 |
+
"epoch": 0.13,
|
| 19 |
+
"learning_rate": 4.8571428571428576e-05,
|
| 20 |
+
"loss": 2.6392,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
"step": 10
|
| 22 |
},
|
| 23 |
{
|
| 24 |
+
"epoch": 0.18,
|
| 25 |
+
"eval_accuracy": 0.25,
|
| 26 |
+
"eval_loss": 2.5779731273651123,
|
| 27 |
+
"eval_runtime": 7.9027,
|
| 28 |
+
"eval_samples_per_second": 3.543,
|
| 29 |
+
"eval_steps_per_second": 0.506,
|
| 30 |
"step": 14
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 1.01,
|
| 34 |
+
"learning_rate": 4.5e-05,
|
| 35 |
+
"loss": 2.5921,
|
| 36 |
"step": 15
|
| 37 |
},
|
| 38 |
{
|
| 39 |
+
"epoch": 1.08,
|
| 40 |
+
"learning_rate": 4.1428571428571437e-05,
|
| 41 |
+
"loss": 2.2172,
|
| 42 |
"step": 20
|
| 43 |
},
|
| 44 |
{
|
| 45 |
+
"epoch": 1.14,
|
| 46 |
+
"learning_rate": 3.785714285714286e-05,
|
| 47 |
+
"loss": 1.994,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
"step": 25
|
| 49 |
},
|
| 50 |
{
|
| 51 |
+
"epoch": 1.18,
|
| 52 |
+
"eval_accuracy": 0.39285714285714285,
|
| 53 |
+
"eval_loss": 2.399493932723999,
|
| 54 |
+
"eval_runtime": 7.938,
|
| 55 |
+
"eval_samples_per_second": 3.527,
|
| 56 |
+
"eval_steps_per_second": 0.504,
|
| 57 |
"step": 28
|
| 58 |
},
|
| 59 |
{
|
| 60 |
+
"epoch": 2.03,
|
| 61 |
+
"learning_rate": 3.428571428571429e-05,
|
| 62 |
+
"loss": 1.947,
|
| 63 |
"step": 30
|
| 64 |
},
|
| 65 |
{
|
| 66 |
+
"epoch": 2.09,
|
| 67 |
+
"learning_rate": 3.071428571428572e-05,
|
| 68 |
+
"loss": 1.4849,
|
| 69 |
"step": 35
|
| 70 |
},
|
| 71 |
{
|
| 72 |
+
"epoch": 2.15,
|
| 73 |
+
"learning_rate": 2.714285714285714e-05,
|
| 74 |
+
"loss": 1.6374,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 75 |
"step": 40
|
| 76 |
},
|
| 77 |
{
|
| 78 |
+
"epoch": 2.18,
|
| 79 |
+
"eval_accuracy": 0.39285714285714285,
|
| 80 |
+
"eval_loss": 2.3010246753692627,
|
| 81 |
+
"eval_runtime": 7.9782,
|
| 82 |
+
"eval_samples_per_second": 3.51,
|
| 83 |
+
"eval_steps_per_second": 0.501,
|
| 84 |
"step": 42
|
| 85 |
},
|
| 86 |
{
|
| 87 |
+
"epoch": 3.04,
|
| 88 |
+
"learning_rate": 2.357142857142857e-05,
|
| 89 |
+
"loss": 1.3647,
|
| 90 |
"step": 45
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"epoch": 3.1,
|
| 94 |
+
"learning_rate": 2e-05,
|
| 95 |
+
"loss": 1.216,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 96 |
"step": 50
|
| 97 |
},
|
| 98 |
{
|
| 99 |
+
"epoch": 3.17,
|
| 100 |
+
"learning_rate": 1.642857142857143e-05,
|
| 101 |
+
"loss": 1.124,
|
| 102 |
"step": 55
|
| 103 |
},
|
| 104 |
{
|
| 105 |
+
"epoch": 3.18,
|
| 106 |
+
"eval_accuracy": 0.39285714285714285,
|
| 107 |
+
"eval_loss": 2.2242484092712402,
|
| 108 |
+
"eval_runtime": 7.7889,
|
| 109 |
+
"eval_samples_per_second": 3.595,
|
| 110 |
+
"eval_steps_per_second": 0.514,
|
| 111 |
"step": 56
|
| 112 |
},
|
| 113 |
{
|
| 114 |
+
"epoch": 4.05,
|
| 115 |
+
"learning_rate": 1.2857142857142857e-05,
|
| 116 |
+
"loss": 1.0848,
|
| 117 |
"step": 60
|
| 118 |
},
|
| 119 |
{
|
| 120 |
+
"epoch": 4.12,
|
| 121 |
+
"learning_rate": 9.285714285714286e-06,
|
| 122 |
+
"loss": 0.9794,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 123 |
"step": 65
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 4.18,
|
| 127 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 128 |
+
"loss": 0.9569,
|
| 129 |
"step": 70
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 4.18,
|
| 133 |
"eval_accuracy": 0.42857142857142855,
|
| 134 |
+
"eval_loss": 2.182518482208252,
|
| 135 |
+
"eval_runtime": 7.9247,
|
| 136 |
+
"eval_samples_per_second": 3.533,
|
| 137 |
+
"eval_steps_per_second": 0.505,
|
| 138 |
"step": 70
|
| 139 |
},
|
| 140 |
{
|
| 141 |
+
"epoch": 5.06,
|
| 142 |
+
"learning_rate": 2.142857142857143e-06,
|
| 143 |
+
"loss": 0.8862,
|
| 144 |
"step": 75
|
| 145 |
},
|
| 146 |
{
|
| 147 |
+
"epoch": 5.1,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 148 |
"eval_accuracy": 0.42857142857142855,
|
| 149 |
+
"eval_loss": 2.1734211444854736,
|
| 150 |
+
"eval_runtime": 7.8943,
|
| 151 |
+
"eval_samples_per_second": 3.547,
|
| 152 |
+
"eval_steps_per_second": 0.507,
|
| 153 |
+
"step": 78
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"epoch": 5.1,
|
| 157 |
+
"step": 78,
|
| 158 |
+
"total_flos": 7.713965251203564e+17,
|
| 159 |
+
"train_loss": 1.6192821661631267,
|
| 160 |
+
"train_runtime": 267.6762,
|
| 161 |
+
"train_samples_per_second": 2.331,
|
| 162 |
+
"train_steps_per_second": 0.291
|
| 163 |
},
|
| 164 |
{
|
| 165 |
+
"epoch": 5.1,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 166 |
"eval_accuracy": 0.42857142857142855,
|
| 167 |
+
"eval_loss": 2.0714259147644043,
|
| 168 |
+
"eval_runtime": 4.1878,
|
| 169 |
+
"eval_samples_per_second": 3.343,
|
| 170 |
+
"eval_steps_per_second": 0.478,
|
| 171 |
+
"step": 78
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 5.1,
|
| 175 |
"eval_accuracy": 0.42857142857142855,
|
| 176 |
+
"eval_loss": 2.0714259147644043,
|
| 177 |
+
"eval_runtime": 3.8707,
|
| 178 |
+
"eval_samples_per_second": 3.617,
|
| 179 |
+
"eval_steps_per_second": 0.517,
|
| 180 |
+
"step": 78
|
| 181 |
}
|
| 182 |
],
|
| 183 |
"logging_steps": 5,
|
| 184 |
+
"max_steps": 78,
|
| 185 |
"num_input_tokens_seen": 0,
|
| 186 |
"num_train_epochs": 9223372036854775807,
|
| 187 |
"save_steps": 500,
|
| 188 |
+
"total_flos": 7.713965251203564e+17,
|
| 189 |
+
"train_batch_size": 8,
|
| 190 |
"trial_name": null,
|
| 191 |
"trial_params": null
|
| 192 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4728
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8a9df7204865a72646bb05717e94fcd96b091033787df0920383d9f482c0253e
|
| 3 |
size 4728
|