Upload folder using huggingface_hub
Browse files- README.md +6 -6
- all_results.json +5 -5
- checkpoint-100/config.json +50 -0
- checkpoint-100/model.safetensors +3 -0
- checkpoint-100/optimizer.pt +3 -0
- checkpoint-100/preprocessor_config.json +42 -0
- checkpoint-100/rng_state.pth +3 -0
- checkpoint-100/scheduler.pt +3 -0
- checkpoint-100/trainer_state.json +109 -0
- checkpoint-100/training_args.bin +3 -0
- checkpoint-150/config.json +50 -0
- checkpoint-150/model.safetensors +3 -0
- checkpoint-150/optimizer.pt +3 -0
- checkpoint-150/preprocessor_config.json +42 -0
- checkpoint-150/rng_state.pth +3 -0
- checkpoint-150/scheduler.pt +3 -0
- checkpoint-150/trainer_state.json +153 -0
- checkpoint-150/training_args.bin +3 -0
- checkpoint-200/config.json +50 -0
- checkpoint-200/model.safetensors +3 -0
- checkpoint-200/optimizer.pt +3 -0
- checkpoint-200/preprocessor_config.json +42 -0
- checkpoint-200/rng_state.pth +3 -0
- checkpoint-200/scheduler.pt +3 -0
- checkpoint-200/trainer_state.json +197 -0
- checkpoint-200/training_args.bin +3 -0
- checkpoint-50/config.json +50 -0
- checkpoint-50/model.safetensors +3 -0
- checkpoint-50/optimizer.pt +3 -0
- checkpoint-50/preprocessor_config.json +42 -0
- checkpoint-50/rng_state.pth +3 -0
- checkpoint-50/scheduler.pt +3 -0
- checkpoint-50/trainer_state.json +65 -0
- checkpoint-50/training_args.bin +3 -0
- model.safetensors +1 -1
- test_results.json +5 -5
- trainer_state.json +73 -73
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [facebook/timesformer-base-finetuned-k400](https://huggingface.co/facebook/timesformer-base-finetuned-k400) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Accuracy: 0
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -50,10 +50,10 @@ The following hyperparameters were used during training:
|
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 52 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 53 |
-
| 0.
|
| 54 |
-
| 0.
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
|
| 58 |
|
| 59 |
### Framework versions
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [facebook/timesformer-base-finetuned-k400](https://huggingface.co/facebook/timesformer-base-finetuned-k400) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.0878
|
| 21 |
+
- Accuracy: 1.0
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 52 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 53 |
+
| 0.1523 | 0.25 | 50 | 0.0735 | 1.0 |
|
| 54 |
+
| 0.024 | 1.25 | 100 | 0.0470 | 0.9866 |
|
| 55 |
+
| 0.0583 | 2.25 | 150 | 0.0302 | 0.9866 |
|
| 56 |
+
| 0.0036 | 3.25 | 200 | 0.0297 | 0.9866 |
|
| 57 |
|
| 58 |
|
| 59 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 3.25,
|
| 3 |
-
"eval_accuracy": 0
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 4.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 3.25,
|
| 3 |
+
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_loss": 0.08780594170093536,
|
| 5 |
+
"eval_runtime": 76.2242,
|
| 6 |
+
"eval_samples_per_second": 4.552,
|
| 7 |
+
"eval_steps_per_second": 0.761
|
| 8 |
}
|
checkpoint-100/config.json
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/timesformer-base-finetuned-k400",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"TimesformerForVideoClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"attention_type": "divided_space_time",
|
| 8 |
+
"drop_path_rate": 0,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.0,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "ApplyEyeMakeup",
|
| 14 |
+
"1": "ApplyLipstick",
|
| 15 |
+
"2": "Archery",
|
| 16 |
+
"3": "BabyCrawling",
|
| 17 |
+
"4": "BalanceBeam",
|
| 18 |
+
"5": "BandMarching",
|
| 19 |
+
"6": "BaseballPitch",
|
| 20 |
+
"7": "Basketball",
|
| 21 |
+
"8": "BasketballDunk",
|
| 22 |
+
"9": "BenchPress"
|
| 23 |
+
},
|
| 24 |
+
"image_size": 224,
|
| 25 |
+
"initializer_range": 0.02,
|
| 26 |
+
"intermediate_size": 3072,
|
| 27 |
+
"label2id": {
|
| 28 |
+
"ApplyEyeMakeup": 0,
|
| 29 |
+
"ApplyLipstick": 1,
|
| 30 |
+
"Archery": 2,
|
| 31 |
+
"BabyCrawling": 3,
|
| 32 |
+
"BalanceBeam": 4,
|
| 33 |
+
"BandMarching": 5,
|
| 34 |
+
"BaseballPitch": 6,
|
| 35 |
+
"Basketball": 7,
|
| 36 |
+
"BasketballDunk": 8,
|
| 37 |
+
"BenchPress": 9
|
| 38 |
+
},
|
| 39 |
+
"layer_norm_eps": 1e-06,
|
| 40 |
+
"model_type": "timesformer",
|
| 41 |
+
"num_attention_heads": 12,
|
| 42 |
+
"num_channels": 3,
|
| 43 |
+
"num_frames": 8,
|
| 44 |
+
"num_hidden_layers": 12,
|
| 45 |
+
"patch_size": 16,
|
| 46 |
+
"problem_type": "single_label_classification",
|
| 47 |
+
"qkv_bias": true,
|
| 48 |
+
"torch_dtype": "float32",
|
| 49 |
+
"transformers_version": "4.40.2"
|
| 50 |
+
}
|
checkpoint-100/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f58ddae36cfa84a5131a3a34ee8e1cd27a8491b7c2359567e46495833d871ab0
|
| 3 |
+
size 485096872
|
checkpoint-100/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0791dbacea186524070a4d3a7948a24254e216b037babbbe0206a2187aa89da5
|
| 3 |
+
size 970332602
|
checkpoint-100/preprocessor_config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"videos",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_center_crop",
|
| 8 |
+
"crop_size",
|
| 9 |
+
"do_rescale",
|
| 10 |
+
"rescale_factor",
|
| 11 |
+
"do_normalize",
|
| 12 |
+
"image_mean",
|
| 13 |
+
"image_std",
|
| 14 |
+
"return_tensors",
|
| 15 |
+
"data_format",
|
| 16 |
+
"input_data_format"
|
| 17 |
+
],
|
| 18 |
+
"crop_size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
},
|
| 22 |
+
"do_center_crop": true,
|
| 23 |
+
"do_normalize": true,
|
| 24 |
+
"do_rescale": true,
|
| 25 |
+
"do_resize": true,
|
| 26 |
+
"image_mean": [
|
| 27 |
+
0.45,
|
| 28 |
+
0.45,
|
| 29 |
+
0.45
|
| 30 |
+
],
|
| 31 |
+
"image_processor_type": "VideoMAEImageProcessor",
|
| 32 |
+
"image_std": [
|
| 33 |
+
0.225,
|
| 34 |
+
0.225,
|
| 35 |
+
0.225
|
| 36 |
+
],
|
| 37 |
+
"resample": 2,
|
| 38 |
+
"rescale_factor": 0.00392156862745098,
|
| 39 |
+
"size": {
|
| 40 |
+
"shortest_edge": 224
|
| 41 |
+
}
|
| 42 |
+
}
|
checkpoint-100/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b67330906a59a9062befee681a594275ad9dcb4b73011e53952685ae3335f37
|
| 3 |
+
size 13990
|
checkpoint-100/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3a2451f1f3a303aa71aa3673eb3a9126da2e85ecf015b17e3e37a7fdfa9c62b5
|
| 3 |
+
size 1064
|
checkpoint-100/trainer_state.json
ADDED
|
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.0,
|
| 3 |
+
"best_model_checkpoint": "timesformer-base-finetuned-k400-finetuned-ucf101-subset/checkpoint-50",
|
| 4 |
+
"epoch": 1.25,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 100,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05,
|
| 13 |
+
"grad_norm": 14.361767768859863,
|
| 14 |
+
"learning_rate": 2.5e-05,
|
| 15 |
+
"loss": 2.3816,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1,
|
| 20 |
+
"grad_norm": 8.448980331420898,
|
| 21 |
+
"learning_rate": 5e-05,
|
| 22 |
+
"loss": 1.5686,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.15,
|
| 27 |
+
"grad_norm": 7.0012526512146,
|
| 28 |
+
"learning_rate": 4.722222222222222e-05,
|
| 29 |
+
"loss": 0.6191,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.2,
|
| 34 |
+
"grad_norm": 4.961479187011719,
|
| 35 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 36 |
+
"loss": 0.3468,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.25,
|
| 41 |
+
"grad_norm": 2.13403058052063,
|
| 42 |
+
"learning_rate": 4.166666666666667e-05,
|
| 43 |
+
"loss": 0.1523,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.25,
|
| 48 |
+
"eval_accuracy": 1.0,
|
| 49 |
+
"eval_loss": 0.07354684174060822,
|
| 50 |
+
"eval_runtime": 34.9344,
|
| 51 |
+
"eval_samples_per_second": 4.265,
|
| 52 |
+
"eval_steps_per_second": 0.716,
|
| 53 |
+
"step": 50
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.05,
|
| 57 |
+
"grad_norm": 0.933645486831665,
|
| 58 |
+
"learning_rate": 3.888888888888889e-05,
|
| 59 |
+
"loss": 0.0521,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.1,
|
| 64 |
+
"grad_norm": 0.865109920501709,
|
| 65 |
+
"learning_rate": 3.611111111111111e-05,
|
| 66 |
+
"loss": 0.0226,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.15,
|
| 71 |
+
"grad_norm": 0.17869924008846283,
|
| 72 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 73 |
+
"loss": 0.108,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.2,
|
| 78 |
+
"grad_norm": 0.23531177639961243,
|
| 79 |
+
"learning_rate": 3.055555555555556e-05,
|
| 80 |
+
"loss": 0.0107,
|
| 81 |
+
"step": 90
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.25,
|
| 85 |
+
"grad_norm": 0.1640271246433258,
|
| 86 |
+
"learning_rate": 2.777777777777778e-05,
|
| 87 |
+
"loss": 0.024,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.25,
|
| 92 |
+
"eval_accuracy": 0.9865771812080537,
|
| 93 |
+
"eval_loss": 0.046999599784612656,
|
| 94 |
+
"eval_runtime": 32.3832,
|
| 95 |
+
"eval_samples_per_second": 4.601,
|
| 96 |
+
"eval_steps_per_second": 0.772,
|
| 97 |
+
"step": 100
|
| 98 |
+
}
|
| 99 |
+
],
|
| 100 |
+
"logging_steps": 10,
|
| 101 |
+
"max_steps": 200,
|
| 102 |
+
"num_input_tokens_seen": 0,
|
| 103 |
+
"num_train_epochs": 9223372036854775807,
|
| 104 |
+
"save_steps": 500,
|
| 105 |
+
"total_flos": 5.257150554636288e+17,
|
| 106 |
+
"train_batch_size": 6,
|
| 107 |
+
"trial_name": null,
|
| 108 |
+
"trial_params": null
|
| 109 |
+
}
|
checkpoint-100/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0976022ff1e3e4b6c6ed3d4b9f4fbf236be50b91b6fc1edad006409b3e04fd11
|
| 3 |
+
size 5048
|
checkpoint-150/config.json
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/timesformer-base-finetuned-k400",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"TimesformerForVideoClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"attention_type": "divided_space_time",
|
| 8 |
+
"drop_path_rate": 0,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.0,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "ApplyEyeMakeup",
|
| 14 |
+
"1": "ApplyLipstick",
|
| 15 |
+
"2": "Archery",
|
| 16 |
+
"3": "BabyCrawling",
|
| 17 |
+
"4": "BalanceBeam",
|
| 18 |
+
"5": "BandMarching",
|
| 19 |
+
"6": "BaseballPitch",
|
| 20 |
+
"7": "Basketball",
|
| 21 |
+
"8": "BasketballDunk",
|
| 22 |
+
"9": "BenchPress"
|
| 23 |
+
},
|
| 24 |
+
"image_size": 224,
|
| 25 |
+
"initializer_range": 0.02,
|
| 26 |
+
"intermediate_size": 3072,
|
| 27 |
+
"label2id": {
|
| 28 |
+
"ApplyEyeMakeup": 0,
|
| 29 |
+
"ApplyLipstick": 1,
|
| 30 |
+
"Archery": 2,
|
| 31 |
+
"BabyCrawling": 3,
|
| 32 |
+
"BalanceBeam": 4,
|
| 33 |
+
"BandMarching": 5,
|
| 34 |
+
"BaseballPitch": 6,
|
| 35 |
+
"Basketball": 7,
|
| 36 |
+
"BasketballDunk": 8,
|
| 37 |
+
"BenchPress": 9
|
| 38 |
+
},
|
| 39 |
+
"layer_norm_eps": 1e-06,
|
| 40 |
+
"model_type": "timesformer",
|
| 41 |
+
"num_attention_heads": 12,
|
| 42 |
+
"num_channels": 3,
|
| 43 |
+
"num_frames": 8,
|
| 44 |
+
"num_hidden_layers": 12,
|
| 45 |
+
"patch_size": 16,
|
| 46 |
+
"problem_type": "single_label_classification",
|
| 47 |
+
"qkv_bias": true,
|
| 48 |
+
"torch_dtype": "float32",
|
| 49 |
+
"transformers_version": "4.40.2"
|
| 50 |
+
}
|
checkpoint-150/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:caf2e2d825cf9d83f687e8d5fa65b487d13187004ae68cc2fd5cab14cfd97e56
|
| 3 |
+
size 485096872
|
checkpoint-150/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:37beaa4e9fa65debae16e635cd70f46879258d9df3abea06f758bebf77a39c7e
|
| 3 |
+
size 970332602
|
checkpoint-150/preprocessor_config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"videos",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_center_crop",
|
| 8 |
+
"crop_size",
|
| 9 |
+
"do_rescale",
|
| 10 |
+
"rescale_factor",
|
| 11 |
+
"do_normalize",
|
| 12 |
+
"image_mean",
|
| 13 |
+
"image_std",
|
| 14 |
+
"return_tensors",
|
| 15 |
+
"data_format",
|
| 16 |
+
"input_data_format"
|
| 17 |
+
],
|
| 18 |
+
"crop_size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
},
|
| 22 |
+
"do_center_crop": true,
|
| 23 |
+
"do_normalize": true,
|
| 24 |
+
"do_rescale": true,
|
| 25 |
+
"do_resize": true,
|
| 26 |
+
"image_mean": [
|
| 27 |
+
0.45,
|
| 28 |
+
0.45,
|
| 29 |
+
0.45
|
| 30 |
+
],
|
| 31 |
+
"image_processor_type": "VideoMAEImageProcessor",
|
| 32 |
+
"image_std": [
|
| 33 |
+
0.225,
|
| 34 |
+
0.225,
|
| 35 |
+
0.225
|
| 36 |
+
],
|
| 37 |
+
"resample": 2,
|
| 38 |
+
"rescale_factor": 0.00392156862745098,
|
| 39 |
+
"size": {
|
| 40 |
+
"shortest_edge": 224
|
| 41 |
+
}
|
| 42 |
+
}
|
checkpoint-150/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f0baa472db4d688b1f9ab6df058c14ddea952a9e067650bddc960b4afd6e9b2b
|
| 3 |
+
size 13990
|
checkpoint-150/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d8f6dcb5f9cb470243f58648097361f079a2dab641af2a564d5ac90bfa5bf025
|
| 3 |
+
size 1064
|
checkpoint-150/trainer_state.json
ADDED
|
@@ -0,0 +1,153 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.0,
|
| 3 |
+
"best_model_checkpoint": "timesformer-base-finetuned-k400-finetuned-ucf101-subset/checkpoint-50",
|
| 4 |
+
"epoch": 2.25,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 150,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05,
|
| 13 |
+
"grad_norm": 14.361767768859863,
|
| 14 |
+
"learning_rate": 2.5e-05,
|
| 15 |
+
"loss": 2.3816,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1,
|
| 20 |
+
"grad_norm": 8.448980331420898,
|
| 21 |
+
"learning_rate": 5e-05,
|
| 22 |
+
"loss": 1.5686,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.15,
|
| 27 |
+
"grad_norm": 7.0012526512146,
|
| 28 |
+
"learning_rate": 4.722222222222222e-05,
|
| 29 |
+
"loss": 0.6191,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.2,
|
| 34 |
+
"grad_norm": 4.961479187011719,
|
| 35 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 36 |
+
"loss": 0.3468,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.25,
|
| 41 |
+
"grad_norm": 2.13403058052063,
|
| 42 |
+
"learning_rate": 4.166666666666667e-05,
|
| 43 |
+
"loss": 0.1523,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.25,
|
| 48 |
+
"eval_accuracy": 1.0,
|
| 49 |
+
"eval_loss": 0.07354684174060822,
|
| 50 |
+
"eval_runtime": 34.9344,
|
| 51 |
+
"eval_samples_per_second": 4.265,
|
| 52 |
+
"eval_steps_per_second": 0.716,
|
| 53 |
+
"step": 50
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.05,
|
| 57 |
+
"grad_norm": 0.933645486831665,
|
| 58 |
+
"learning_rate": 3.888888888888889e-05,
|
| 59 |
+
"loss": 0.0521,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.1,
|
| 64 |
+
"grad_norm": 0.865109920501709,
|
| 65 |
+
"learning_rate": 3.611111111111111e-05,
|
| 66 |
+
"loss": 0.0226,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.15,
|
| 71 |
+
"grad_norm": 0.17869924008846283,
|
| 72 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 73 |
+
"loss": 0.108,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.2,
|
| 78 |
+
"grad_norm": 0.23531177639961243,
|
| 79 |
+
"learning_rate": 3.055555555555556e-05,
|
| 80 |
+
"loss": 0.0107,
|
| 81 |
+
"step": 90
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.25,
|
| 85 |
+
"grad_norm": 0.1640271246433258,
|
| 86 |
+
"learning_rate": 2.777777777777778e-05,
|
| 87 |
+
"loss": 0.024,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.25,
|
| 92 |
+
"eval_accuracy": 0.9865771812080537,
|
| 93 |
+
"eval_loss": 0.046999599784612656,
|
| 94 |
+
"eval_runtime": 32.3832,
|
| 95 |
+
"eval_samples_per_second": 4.601,
|
| 96 |
+
"eval_steps_per_second": 0.772,
|
| 97 |
+
"step": 100
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.05,
|
| 101 |
+
"grad_norm": 0.47139155864715576,
|
| 102 |
+
"learning_rate": 2.5e-05,
|
| 103 |
+
"loss": 0.0056,
|
| 104 |
+
"step": 110
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.1,
|
| 108 |
+
"grad_norm": 0.11248086392879486,
|
| 109 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 110 |
+
"loss": 0.0045,
|
| 111 |
+
"step": 120
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.15,
|
| 115 |
+
"grad_norm": 0.07324650138616562,
|
| 116 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 117 |
+
"loss": 0.0042,
|
| 118 |
+
"step": 130
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.2,
|
| 122 |
+
"grad_norm": 0.46895766258239746,
|
| 123 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 124 |
+
"loss": 0.0202,
|
| 125 |
+
"step": 140
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.25,
|
| 129 |
+
"grad_norm": 0.048765964806079865,
|
| 130 |
+
"learning_rate": 1.388888888888889e-05,
|
| 131 |
+
"loss": 0.0583,
|
| 132 |
+
"step": 150
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 2.25,
|
| 136 |
+
"eval_accuracy": 0.9865771812080537,
|
| 137 |
+
"eval_loss": 0.030154313892126083,
|
| 138 |
+
"eval_runtime": 32.7398,
|
| 139 |
+
"eval_samples_per_second": 4.551,
|
| 140 |
+
"eval_steps_per_second": 0.764,
|
| 141 |
+
"step": 150
|
| 142 |
+
}
|
| 143 |
+
],
|
| 144 |
+
"logging_steps": 10,
|
| 145 |
+
"max_steps": 200,
|
| 146 |
+
"num_input_tokens_seen": 0,
|
| 147 |
+
"num_train_epochs": 9223372036854775807,
|
| 148 |
+
"save_steps": 500,
|
| 149 |
+
"total_flos": 7.885725831954432e+17,
|
| 150 |
+
"train_batch_size": 6,
|
| 151 |
+
"trial_name": null,
|
| 152 |
+
"trial_params": null
|
| 153 |
+
}
|
checkpoint-150/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0976022ff1e3e4b6c6ed3d4b9f4fbf236be50b91b6fc1edad006409b3e04fd11
|
| 3 |
+
size 5048
|
checkpoint-200/config.json
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/timesformer-base-finetuned-k400",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"TimesformerForVideoClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"attention_type": "divided_space_time",
|
| 8 |
+
"drop_path_rate": 0,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.0,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "ApplyEyeMakeup",
|
| 14 |
+
"1": "ApplyLipstick",
|
| 15 |
+
"2": "Archery",
|
| 16 |
+
"3": "BabyCrawling",
|
| 17 |
+
"4": "BalanceBeam",
|
| 18 |
+
"5": "BandMarching",
|
| 19 |
+
"6": "BaseballPitch",
|
| 20 |
+
"7": "Basketball",
|
| 21 |
+
"8": "BasketballDunk",
|
| 22 |
+
"9": "BenchPress"
|
| 23 |
+
},
|
| 24 |
+
"image_size": 224,
|
| 25 |
+
"initializer_range": 0.02,
|
| 26 |
+
"intermediate_size": 3072,
|
| 27 |
+
"label2id": {
|
| 28 |
+
"ApplyEyeMakeup": 0,
|
| 29 |
+
"ApplyLipstick": 1,
|
| 30 |
+
"Archery": 2,
|
| 31 |
+
"BabyCrawling": 3,
|
| 32 |
+
"BalanceBeam": 4,
|
| 33 |
+
"BandMarching": 5,
|
| 34 |
+
"BaseballPitch": 6,
|
| 35 |
+
"Basketball": 7,
|
| 36 |
+
"BasketballDunk": 8,
|
| 37 |
+
"BenchPress": 9
|
| 38 |
+
},
|
| 39 |
+
"layer_norm_eps": 1e-06,
|
| 40 |
+
"model_type": "timesformer",
|
| 41 |
+
"num_attention_heads": 12,
|
| 42 |
+
"num_channels": 3,
|
| 43 |
+
"num_frames": 8,
|
| 44 |
+
"num_hidden_layers": 12,
|
| 45 |
+
"patch_size": 16,
|
| 46 |
+
"problem_type": "single_label_classification",
|
| 47 |
+
"qkv_bias": true,
|
| 48 |
+
"torch_dtype": "float32",
|
| 49 |
+
"transformers_version": "4.40.2"
|
| 50 |
+
}
|
checkpoint-200/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1cdb45c2c1b530a192bdaaf7b195f81fcd649cded3580d8dd7d95e745c3a524a
|
| 3 |
+
size 485096872
|
checkpoint-200/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc6b66fbe3dbe95cc5fa386cad63563ea695fec8b64c350c6739d668d582b54a
|
| 3 |
+
size 970332602
|
checkpoint-200/preprocessor_config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"videos",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_center_crop",
|
| 8 |
+
"crop_size",
|
| 9 |
+
"do_rescale",
|
| 10 |
+
"rescale_factor",
|
| 11 |
+
"do_normalize",
|
| 12 |
+
"image_mean",
|
| 13 |
+
"image_std",
|
| 14 |
+
"return_tensors",
|
| 15 |
+
"data_format",
|
| 16 |
+
"input_data_format"
|
| 17 |
+
],
|
| 18 |
+
"crop_size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
},
|
| 22 |
+
"do_center_crop": true,
|
| 23 |
+
"do_normalize": true,
|
| 24 |
+
"do_rescale": true,
|
| 25 |
+
"do_resize": true,
|
| 26 |
+
"image_mean": [
|
| 27 |
+
0.45,
|
| 28 |
+
0.45,
|
| 29 |
+
0.45
|
| 30 |
+
],
|
| 31 |
+
"image_processor_type": "VideoMAEImageProcessor",
|
| 32 |
+
"image_std": [
|
| 33 |
+
0.225,
|
| 34 |
+
0.225,
|
| 35 |
+
0.225
|
| 36 |
+
],
|
| 37 |
+
"resample": 2,
|
| 38 |
+
"rescale_factor": 0.00392156862745098,
|
| 39 |
+
"size": {
|
| 40 |
+
"shortest_edge": 224
|
| 41 |
+
}
|
| 42 |
+
}
|
checkpoint-200/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:148c967811f2bd1e6262048e250f40991c0da1be099be3b3182978be32de18ed
|
| 3 |
+
size 13990
|
checkpoint-200/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a27b052646dcb561cbd68156c30bd466ce59bda64cf3c8eba9c3c1113af9827c
|
| 3 |
+
size 1064
|
checkpoint-200/trainer_state.json
ADDED
|
@@ -0,0 +1,197 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.0,
|
| 3 |
+
"best_model_checkpoint": "timesformer-base-finetuned-k400-finetuned-ucf101-subset/checkpoint-50",
|
| 4 |
+
"epoch": 3.25,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 200,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05,
|
| 13 |
+
"grad_norm": 14.361767768859863,
|
| 14 |
+
"learning_rate": 2.5e-05,
|
| 15 |
+
"loss": 2.3816,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1,
|
| 20 |
+
"grad_norm": 8.448980331420898,
|
| 21 |
+
"learning_rate": 5e-05,
|
| 22 |
+
"loss": 1.5686,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.15,
|
| 27 |
+
"grad_norm": 7.0012526512146,
|
| 28 |
+
"learning_rate": 4.722222222222222e-05,
|
| 29 |
+
"loss": 0.6191,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.2,
|
| 34 |
+
"grad_norm": 4.961479187011719,
|
| 35 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 36 |
+
"loss": 0.3468,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.25,
|
| 41 |
+
"grad_norm": 2.13403058052063,
|
| 42 |
+
"learning_rate": 4.166666666666667e-05,
|
| 43 |
+
"loss": 0.1523,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.25,
|
| 48 |
+
"eval_accuracy": 1.0,
|
| 49 |
+
"eval_loss": 0.07354684174060822,
|
| 50 |
+
"eval_runtime": 34.9344,
|
| 51 |
+
"eval_samples_per_second": 4.265,
|
| 52 |
+
"eval_steps_per_second": 0.716,
|
| 53 |
+
"step": 50
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.05,
|
| 57 |
+
"grad_norm": 0.933645486831665,
|
| 58 |
+
"learning_rate": 3.888888888888889e-05,
|
| 59 |
+
"loss": 0.0521,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.1,
|
| 64 |
+
"grad_norm": 0.865109920501709,
|
| 65 |
+
"learning_rate": 3.611111111111111e-05,
|
| 66 |
+
"loss": 0.0226,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.15,
|
| 71 |
+
"grad_norm": 0.17869924008846283,
|
| 72 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 73 |
+
"loss": 0.108,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.2,
|
| 78 |
+
"grad_norm": 0.23531177639961243,
|
| 79 |
+
"learning_rate": 3.055555555555556e-05,
|
| 80 |
+
"loss": 0.0107,
|
| 81 |
+
"step": 90
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.25,
|
| 85 |
+
"grad_norm": 0.1640271246433258,
|
| 86 |
+
"learning_rate": 2.777777777777778e-05,
|
| 87 |
+
"loss": 0.024,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.25,
|
| 92 |
+
"eval_accuracy": 0.9865771812080537,
|
| 93 |
+
"eval_loss": 0.046999599784612656,
|
| 94 |
+
"eval_runtime": 32.3832,
|
| 95 |
+
"eval_samples_per_second": 4.601,
|
| 96 |
+
"eval_steps_per_second": 0.772,
|
| 97 |
+
"step": 100
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.05,
|
| 101 |
+
"grad_norm": 0.47139155864715576,
|
| 102 |
+
"learning_rate": 2.5e-05,
|
| 103 |
+
"loss": 0.0056,
|
| 104 |
+
"step": 110
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.1,
|
| 108 |
+
"grad_norm": 0.11248086392879486,
|
| 109 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 110 |
+
"loss": 0.0045,
|
| 111 |
+
"step": 120
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.15,
|
| 115 |
+
"grad_norm": 0.07324650138616562,
|
| 116 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 117 |
+
"loss": 0.0042,
|
| 118 |
+
"step": 130
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.2,
|
| 122 |
+
"grad_norm": 0.46895766258239746,
|
| 123 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 124 |
+
"loss": 0.0202,
|
| 125 |
+
"step": 140
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.25,
|
| 129 |
+
"grad_norm": 0.048765964806079865,
|
| 130 |
+
"learning_rate": 1.388888888888889e-05,
|
| 131 |
+
"loss": 0.0583,
|
| 132 |
+
"step": 150
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 2.25,
|
| 136 |
+
"eval_accuracy": 0.9865771812080537,
|
| 137 |
+
"eval_loss": 0.030154313892126083,
|
| 138 |
+
"eval_runtime": 32.7398,
|
| 139 |
+
"eval_samples_per_second": 4.551,
|
| 140 |
+
"eval_steps_per_second": 0.764,
|
| 141 |
+
"step": 150
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.05,
|
| 145 |
+
"grad_norm": 0.11046731472015381,
|
| 146 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 147 |
+
"loss": 0.0032,
|
| 148 |
+
"step": 160
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 3.1,
|
| 152 |
+
"grad_norm": 0.07638247311115265,
|
| 153 |
+
"learning_rate": 8.333333333333334e-06,
|
| 154 |
+
"loss": 0.0027,
|
| 155 |
+
"step": 170
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 3.15,
|
| 159 |
+
"grad_norm": 0.0314478874206543,
|
| 160 |
+
"learning_rate": 5.555555555555556e-06,
|
| 161 |
+
"loss": 0.0053,
|
| 162 |
+
"step": 180
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 3.2,
|
| 166 |
+
"grad_norm": 0.23283490538597107,
|
| 167 |
+
"learning_rate": 2.777777777777778e-06,
|
| 168 |
+
"loss": 0.0032,
|
| 169 |
+
"step": 190
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 3.25,
|
| 173 |
+
"grad_norm": 0.3815344274044037,
|
| 174 |
+
"learning_rate": 0.0,
|
| 175 |
+
"loss": 0.0036,
|
| 176 |
+
"step": 200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 3.25,
|
| 180 |
+
"eval_accuracy": 0.9865771812080537,
|
| 181 |
+
"eval_loss": 0.02967873215675354,
|
| 182 |
+
"eval_runtime": 32.6048,
|
| 183 |
+
"eval_samples_per_second": 4.57,
|
| 184 |
+
"eval_steps_per_second": 0.767,
|
| 185 |
+
"step": 200
|
| 186 |
+
}
|
| 187 |
+
],
|
| 188 |
+
"logging_steps": 10,
|
| 189 |
+
"max_steps": 200,
|
| 190 |
+
"num_input_tokens_seen": 0,
|
| 191 |
+
"num_train_epochs": 9223372036854775807,
|
| 192 |
+
"save_steps": 500,
|
| 193 |
+
"total_flos": 1.0514301109272576e+18,
|
| 194 |
+
"train_batch_size": 6,
|
| 195 |
+
"trial_name": null,
|
| 196 |
+
"trial_params": null
|
| 197 |
+
}
|
checkpoint-200/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0976022ff1e3e4b6c6ed3d4b9f4fbf236be50b91b6fc1edad006409b3e04fd11
|
| 3 |
+
size 5048
|
checkpoint-50/config.json
ADDED
|
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/timesformer-base-finetuned-k400",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"TimesformerForVideoClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"attention_type": "divided_space_time",
|
| 8 |
+
"drop_path_rate": 0,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.0,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "ApplyEyeMakeup",
|
| 14 |
+
"1": "ApplyLipstick",
|
| 15 |
+
"2": "Archery",
|
| 16 |
+
"3": "BabyCrawling",
|
| 17 |
+
"4": "BalanceBeam",
|
| 18 |
+
"5": "BandMarching",
|
| 19 |
+
"6": "BaseballPitch",
|
| 20 |
+
"7": "Basketball",
|
| 21 |
+
"8": "BasketballDunk",
|
| 22 |
+
"9": "BenchPress"
|
| 23 |
+
},
|
| 24 |
+
"image_size": 224,
|
| 25 |
+
"initializer_range": 0.02,
|
| 26 |
+
"intermediate_size": 3072,
|
| 27 |
+
"label2id": {
|
| 28 |
+
"ApplyEyeMakeup": 0,
|
| 29 |
+
"ApplyLipstick": 1,
|
| 30 |
+
"Archery": 2,
|
| 31 |
+
"BabyCrawling": 3,
|
| 32 |
+
"BalanceBeam": 4,
|
| 33 |
+
"BandMarching": 5,
|
| 34 |
+
"BaseballPitch": 6,
|
| 35 |
+
"Basketball": 7,
|
| 36 |
+
"BasketballDunk": 8,
|
| 37 |
+
"BenchPress": 9
|
| 38 |
+
},
|
| 39 |
+
"layer_norm_eps": 1e-06,
|
| 40 |
+
"model_type": "timesformer",
|
| 41 |
+
"num_attention_heads": 12,
|
| 42 |
+
"num_channels": 3,
|
| 43 |
+
"num_frames": 8,
|
| 44 |
+
"num_hidden_layers": 12,
|
| 45 |
+
"patch_size": 16,
|
| 46 |
+
"problem_type": "single_label_classification",
|
| 47 |
+
"qkv_bias": true,
|
| 48 |
+
"torch_dtype": "float32",
|
| 49 |
+
"transformers_version": "4.40.2"
|
| 50 |
+
}
|
checkpoint-50/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9376ab61570e92ddbc75657778a029c7977423294333a9d8b7544c679041f40c
|
| 3 |
+
size 485096872
|
checkpoint-50/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b7c67ffd16e3925d25befc1362eec1f24d96f2ede12042a6739a9ff6fa861000
|
| 3 |
+
size 970332602
|
checkpoint-50/preprocessor_config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"videos",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_center_crop",
|
| 8 |
+
"crop_size",
|
| 9 |
+
"do_rescale",
|
| 10 |
+
"rescale_factor",
|
| 11 |
+
"do_normalize",
|
| 12 |
+
"image_mean",
|
| 13 |
+
"image_std",
|
| 14 |
+
"return_tensors",
|
| 15 |
+
"data_format",
|
| 16 |
+
"input_data_format"
|
| 17 |
+
],
|
| 18 |
+
"crop_size": {
|
| 19 |
+
"height": 224,
|
| 20 |
+
"width": 224
|
| 21 |
+
},
|
| 22 |
+
"do_center_crop": true,
|
| 23 |
+
"do_normalize": true,
|
| 24 |
+
"do_rescale": true,
|
| 25 |
+
"do_resize": true,
|
| 26 |
+
"image_mean": [
|
| 27 |
+
0.45,
|
| 28 |
+
0.45,
|
| 29 |
+
0.45
|
| 30 |
+
],
|
| 31 |
+
"image_processor_type": "VideoMAEImageProcessor",
|
| 32 |
+
"image_std": [
|
| 33 |
+
0.225,
|
| 34 |
+
0.225,
|
| 35 |
+
0.225
|
| 36 |
+
],
|
| 37 |
+
"resample": 2,
|
| 38 |
+
"rescale_factor": 0.00392156862745098,
|
| 39 |
+
"size": {
|
| 40 |
+
"shortest_edge": 224
|
| 41 |
+
}
|
| 42 |
+
}
|
checkpoint-50/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b7b9789f20324e67c1622b7a879ebea7dd5465bdcf756c8dc0893766b42a9392
|
| 3 |
+
size 13990
|
checkpoint-50/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:716493262cdbc3f2943f4f3c7d87218de0b57dea1d2f05df1de6cfa59a812dcb
|
| 3 |
+
size 1064
|
checkpoint-50/trainer_state.json
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.0,
|
| 3 |
+
"best_model_checkpoint": "timesformer-base-finetuned-k400-finetuned-ucf101-subset/checkpoint-50",
|
| 4 |
+
"epoch": 0.25,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 50,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05,
|
| 13 |
+
"grad_norm": 14.361767768859863,
|
| 14 |
+
"learning_rate": 2.5e-05,
|
| 15 |
+
"loss": 2.3816,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1,
|
| 20 |
+
"grad_norm": 8.448980331420898,
|
| 21 |
+
"learning_rate": 5e-05,
|
| 22 |
+
"loss": 1.5686,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.15,
|
| 27 |
+
"grad_norm": 7.0012526512146,
|
| 28 |
+
"learning_rate": 4.722222222222222e-05,
|
| 29 |
+
"loss": 0.6191,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.2,
|
| 34 |
+
"grad_norm": 4.961479187011719,
|
| 35 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 36 |
+
"loss": 0.3468,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.25,
|
| 41 |
+
"grad_norm": 2.13403058052063,
|
| 42 |
+
"learning_rate": 4.166666666666667e-05,
|
| 43 |
+
"loss": 0.1523,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.25,
|
| 48 |
+
"eval_accuracy": 1.0,
|
| 49 |
+
"eval_loss": 0.07354684174060822,
|
| 50 |
+
"eval_runtime": 34.9344,
|
| 51 |
+
"eval_samples_per_second": 4.265,
|
| 52 |
+
"eval_steps_per_second": 0.716,
|
| 53 |
+
"step": 50
|
| 54 |
+
}
|
| 55 |
+
],
|
| 56 |
+
"logging_steps": 10,
|
| 57 |
+
"max_steps": 200,
|
| 58 |
+
"num_input_tokens_seen": 0,
|
| 59 |
+
"num_train_epochs": 9223372036854775807,
|
| 60 |
+
"save_steps": 500,
|
| 61 |
+
"total_flos": 2.628575277318144e+17,
|
| 62 |
+
"train_batch_size": 6,
|
| 63 |
+
"trial_name": null,
|
| 64 |
+
"trial_params": null
|
| 65 |
+
}
|
checkpoint-50/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0976022ff1e3e4b6c6ed3d4b9f4fbf236be50b91b6fc1edad006409b3e04fd11
|
| 3 |
+
size 5048
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 485096872
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9376ab61570e92ddbc75657778a029c7977423294333a9d8b7544c679041f40c
|
| 3 |
size 485096872
|
test_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"epoch": 3.25,
|
| 3 |
-
"eval_accuracy": 0
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second": 4.
|
| 7 |
-
"eval_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"epoch": 3.25,
|
| 3 |
+
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_loss": 0.08780594170093536,
|
| 5 |
+
"eval_runtime": 76.2242,
|
| 6 |
+
"eval_samples_per_second": 4.552,
|
| 7 |
+
"eval_steps_per_second": 0.761
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -10,205 +10,205 @@
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.05,
|
| 13 |
-
"grad_norm":
|
| 14 |
"learning_rate": 2.5e-05,
|
| 15 |
-
"loss": 2.
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.1,
|
| 20 |
-
"grad_norm":
|
| 21 |
"learning_rate": 5e-05,
|
| 22 |
-
"loss": 1.
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.15,
|
| 27 |
-
"grad_norm":
|
| 28 |
"learning_rate": 4.722222222222222e-05,
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.2,
|
| 34 |
-
"grad_norm":
|
| 35 |
"learning_rate": 4.4444444444444447e-05,
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.25,
|
| 41 |
-
"grad_norm":
|
| 42 |
"learning_rate": 4.166666666666667e-05,
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.25,
|
| 48 |
"eval_accuracy": 1.0,
|
| 49 |
-
"eval_loss": 0.
|
| 50 |
-
"eval_runtime":
|
| 51 |
-
"eval_samples_per_second": 4.
|
| 52 |
-
"eval_steps_per_second": 0.
|
| 53 |
"step": 50
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 1.05,
|
| 57 |
-
"grad_norm": 0.
|
| 58 |
"learning_rate": 3.888888888888889e-05,
|
| 59 |
-
"loss": 0.
|
| 60 |
"step": 60
|
| 61 |
},
|
| 62 |
{
|
| 63 |
"epoch": 1.1,
|
| 64 |
-
"grad_norm": 0.
|
| 65 |
"learning_rate": 3.611111111111111e-05,
|
| 66 |
-
"loss": 0.
|
| 67 |
"step": 70
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"epoch": 1.15,
|
| 71 |
-
"grad_norm": 0.
|
| 72 |
"learning_rate": 3.3333333333333335e-05,
|
| 73 |
-
"loss": 0.
|
| 74 |
"step": 80
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 1.2,
|
| 78 |
-
"grad_norm": 0.
|
| 79 |
"learning_rate": 3.055555555555556e-05,
|
| 80 |
-
"loss": 0.
|
| 81 |
"step": 90
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 1.25,
|
| 85 |
-
"grad_norm": 0.
|
| 86 |
"learning_rate": 2.777777777777778e-05,
|
| 87 |
-
"loss": 0.
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 1.25,
|
| 92 |
-
"eval_accuracy":
|
| 93 |
-
"eval_loss": 0.
|
| 94 |
-
"eval_runtime": 32.
|
| 95 |
-
"eval_samples_per_second": 4.
|
| 96 |
-
"eval_steps_per_second": 0.
|
| 97 |
"step": 100
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"epoch": 2.05,
|
| 101 |
-
"grad_norm":
|
| 102 |
"learning_rate": 2.5e-05,
|
| 103 |
-
"loss": 0.
|
| 104 |
"step": 110
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 2.1,
|
| 108 |
-
"grad_norm": 0.
|
| 109 |
"learning_rate": 2.2222222222222223e-05,
|
| 110 |
-
"loss": 0.
|
| 111 |
"step": 120
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 2.15,
|
| 115 |
-
"grad_norm": 0.
|
| 116 |
"learning_rate": 1.9444444444444445e-05,
|
| 117 |
-
"loss": 0.
|
| 118 |
"step": 130
|
| 119 |
},
|
| 120 |
{
|
| 121 |
"epoch": 2.2,
|
| 122 |
-
"grad_norm": 0.
|
| 123 |
"learning_rate": 1.6666666666666667e-05,
|
| 124 |
-
"loss": 0.
|
| 125 |
"step": 140
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 2.25,
|
| 129 |
-
"grad_norm": 0.
|
| 130 |
"learning_rate": 1.388888888888889e-05,
|
| 131 |
-
"loss": 0.
|
| 132 |
"step": 150
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 2.25,
|
| 136 |
-
"eval_accuracy":
|
| 137 |
-
"eval_loss": 0.
|
| 138 |
-
"eval_runtime": 32.
|
| 139 |
-
"eval_samples_per_second": 4.
|
| 140 |
-
"eval_steps_per_second": 0.
|
| 141 |
"step": 150
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 3.05,
|
| 145 |
-
"grad_norm": 0.
|
| 146 |
"learning_rate": 1.1111111111111112e-05,
|
| 147 |
-
"loss": 0.
|
| 148 |
"step": 160
|
| 149 |
},
|
| 150 |
{
|
| 151 |
"epoch": 3.1,
|
| 152 |
-
"grad_norm": 0.
|
| 153 |
"learning_rate": 8.333333333333334e-06,
|
| 154 |
-
"loss": 0.
|
| 155 |
"step": 170
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 3.15,
|
| 159 |
-
"grad_norm": 0.
|
| 160 |
"learning_rate": 5.555555555555556e-06,
|
| 161 |
-
"loss": 0.
|
| 162 |
"step": 180
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 3.2,
|
| 166 |
-
"grad_norm": 0.
|
| 167 |
"learning_rate": 2.777777777777778e-06,
|
| 168 |
-
"loss": 0.
|
| 169 |
"step": 190
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 3.25,
|
| 173 |
-
"grad_norm": 0.
|
| 174 |
"learning_rate": 0.0,
|
| 175 |
-
"loss": 0.
|
| 176 |
"step": 200
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 3.25,
|
| 180 |
-
"eval_accuracy": 0.
|
| 181 |
-
"eval_loss": 0.
|
| 182 |
-
"eval_runtime": 32.
|
| 183 |
-
"eval_samples_per_second": 4.
|
| 184 |
-
"eval_steps_per_second": 0.
|
| 185 |
"step": 200
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 3.25,
|
| 189 |
"step": 200,
|
| 190 |
"total_flos": 1.0514301109272576e+18,
|
| 191 |
-
"train_loss": 0.
|
| 192 |
-
"train_runtime":
|
| 193 |
-
"train_samples_per_second": 1.
|
| 194 |
-
"train_steps_per_second": 0.
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 3.25,
|
| 198 |
-
"eval_accuracy": 0
|
| 199 |
-
"eval_loss": 0.
|
| 200 |
-
"eval_runtime":
|
| 201 |
-
"eval_samples_per_second": 4.
|
| 202 |
-
"eval_steps_per_second": 0.
|
| 203 |
"step": 200
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 3.25,
|
| 207 |
-
"eval_accuracy": 0
|
| 208 |
-
"eval_loss": 0.
|
| 209 |
-
"eval_runtime":
|
| 210 |
-
"eval_samples_per_second": 4.
|
| 211 |
-
"eval_steps_per_second": 0.
|
| 212 |
"step": 200
|
| 213 |
}
|
| 214 |
],
|
|
|
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.05,
|
| 13 |
+
"grad_norm": 14.361767768859863,
|
| 14 |
"learning_rate": 2.5e-05,
|
| 15 |
+
"loss": 2.3816,
|
| 16 |
"step": 10
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.1,
|
| 20 |
+
"grad_norm": 8.448980331420898,
|
| 21 |
"learning_rate": 5e-05,
|
| 22 |
+
"loss": 1.5686,
|
| 23 |
"step": 20
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.15,
|
| 27 |
+
"grad_norm": 7.0012526512146,
|
| 28 |
"learning_rate": 4.722222222222222e-05,
|
| 29 |
+
"loss": 0.6191,
|
| 30 |
"step": 30
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.2,
|
| 34 |
+
"grad_norm": 4.961479187011719,
|
| 35 |
"learning_rate": 4.4444444444444447e-05,
|
| 36 |
+
"loss": 0.3468,
|
| 37 |
"step": 40
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.25,
|
| 41 |
+
"grad_norm": 2.13403058052063,
|
| 42 |
"learning_rate": 4.166666666666667e-05,
|
| 43 |
+
"loss": 0.1523,
|
| 44 |
"step": 50
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.25,
|
| 48 |
"eval_accuracy": 1.0,
|
| 49 |
+
"eval_loss": 0.07354684174060822,
|
| 50 |
+
"eval_runtime": 34.9344,
|
| 51 |
+
"eval_samples_per_second": 4.265,
|
| 52 |
+
"eval_steps_per_second": 0.716,
|
| 53 |
"step": 50
|
| 54 |
},
|
| 55 |
{
|
| 56 |
"epoch": 1.05,
|
| 57 |
+
"grad_norm": 0.933645486831665,
|
| 58 |
"learning_rate": 3.888888888888889e-05,
|
| 59 |
+
"loss": 0.0521,
|
| 60 |
"step": 60
|
| 61 |
},
|
| 62 |
{
|
| 63 |
"epoch": 1.1,
|
| 64 |
+
"grad_norm": 0.865109920501709,
|
| 65 |
"learning_rate": 3.611111111111111e-05,
|
| 66 |
+
"loss": 0.0226,
|
| 67 |
"step": 70
|
| 68 |
},
|
| 69 |
{
|
| 70 |
"epoch": 1.15,
|
| 71 |
+
"grad_norm": 0.17869924008846283,
|
| 72 |
"learning_rate": 3.3333333333333335e-05,
|
| 73 |
+
"loss": 0.108,
|
| 74 |
"step": 80
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 1.2,
|
| 78 |
+
"grad_norm": 0.23531177639961243,
|
| 79 |
"learning_rate": 3.055555555555556e-05,
|
| 80 |
+
"loss": 0.0107,
|
| 81 |
"step": 90
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 1.25,
|
| 85 |
+
"grad_norm": 0.1640271246433258,
|
| 86 |
"learning_rate": 2.777777777777778e-05,
|
| 87 |
+
"loss": 0.024,
|
| 88 |
"step": 100
|
| 89 |
},
|
| 90 |
{
|
| 91 |
"epoch": 1.25,
|
| 92 |
+
"eval_accuracy": 0.9865771812080537,
|
| 93 |
+
"eval_loss": 0.046999599784612656,
|
| 94 |
+
"eval_runtime": 32.3832,
|
| 95 |
+
"eval_samples_per_second": 4.601,
|
| 96 |
+
"eval_steps_per_second": 0.772,
|
| 97 |
"step": 100
|
| 98 |
},
|
| 99 |
{
|
| 100 |
"epoch": 2.05,
|
| 101 |
+
"grad_norm": 0.47139155864715576,
|
| 102 |
"learning_rate": 2.5e-05,
|
| 103 |
+
"loss": 0.0056,
|
| 104 |
"step": 110
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 2.1,
|
| 108 |
+
"grad_norm": 0.11248086392879486,
|
| 109 |
"learning_rate": 2.2222222222222223e-05,
|
| 110 |
+
"loss": 0.0045,
|
| 111 |
"step": 120
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 2.15,
|
| 115 |
+
"grad_norm": 0.07324650138616562,
|
| 116 |
"learning_rate": 1.9444444444444445e-05,
|
| 117 |
+
"loss": 0.0042,
|
| 118 |
"step": 130
|
| 119 |
},
|
| 120 |
{
|
| 121 |
"epoch": 2.2,
|
| 122 |
+
"grad_norm": 0.46895766258239746,
|
| 123 |
"learning_rate": 1.6666666666666667e-05,
|
| 124 |
+
"loss": 0.0202,
|
| 125 |
"step": 140
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"epoch": 2.25,
|
| 129 |
+
"grad_norm": 0.048765964806079865,
|
| 130 |
"learning_rate": 1.388888888888889e-05,
|
| 131 |
+
"loss": 0.0583,
|
| 132 |
"step": 150
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 2.25,
|
| 136 |
+
"eval_accuracy": 0.9865771812080537,
|
| 137 |
+
"eval_loss": 0.030154313892126083,
|
| 138 |
+
"eval_runtime": 32.7398,
|
| 139 |
+
"eval_samples_per_second": 4.551,
|
| 140 |
+
"eval_steps_per_second": 0.764,
|
| 141 |
"step": 150
|
| 142 |
},
|
| 143 |
{
|
| 144 |
"epoch": 3.05,
|
| 145 |
+
"grad_norm": 0.11046731472015381,
|
| 146 |
"learning_rate": 1.1111111111111112e-05,
|
| 147 |
+
"loss": 0.0032,
|
| 148 |
"step": 160
|
| 149 |
},
|
| 150 |
{
|
| 151 |
"epoch": 3.1,
|
| 152 |
+
"grad_norm": 0.07638247311115265,
|
| 153 |
"learning_rate": 8.333333333333334e-06,
|
| 154 |
+
"loss": 0.0027,
|
| 155 |
"step": 170
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 3.15,
|
| 159 |
+
"grad_norm": 0.0314478874206543,
|
| 160 |
"learning_rate": 5.555555555555556e-06,
|
| 161 |
+
"loss": 0.0053,
|
| 162 |
"step": 180
|
| 163 |
},
|
| 164 |
{
|
| 165 |
"epoch": 3.2,
|
| 166 |
+
"grad_norm": 0.23283490538597107,
|
| 167 |
"learning_rate": 2.777777777777778e-06,
|
| 168 |
+
"loss": 0.0032,
|
| 169 |
"step": 190
|
| 170 |
},
|
| 171 |
{
|
| 172 |
"epoch": 3.25,
|
| 173 |
+
"grad_norm": 0.3815344274044037,
|
| 174 |
"learning_rate": 0.0,
|
| 175 |
+
"loss": 0.0036,
|
| 176 |
"step": 200
|
| 177 |
},
|
| 178 |
{
|
| 179 |
"epoch": 3.25,
|
| 180 |
+
"eval_accuracy": 0.9865771812080537,
|
| 181 |
+
"eval_loss": 0.02967873215675354,
|
| 182 |
+
"eval_runtime": 32.6048,
|
| 183 |
+
"eval_samples_per_second": 4.57,
|
| 184 |
+
"eval_steps_per_second": 0.767,
|
| 185 |
"step": 200
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 3.25,
|
| 189 |
"step": 200,
|
| 190 |
"total_flos": 1.0514301109272576e+18,
|
| 191 |
+
"train_loss": 0.2698267618194222,
|
| 192 |
+
"train_runtime": 1064.2955,
|
| 193 |
+
"train_samples_per_second": 1.128,
|
| 194 |
+
"train_steps_per_second": 0.188
|
| 195 |
},
|
| 196 |
{
|
| 197 |
"epoch": 3.25,
|
| 198 |
+
"eval_accuracy": 1.0,
|
| 199 |
+
"eval_loss": 0.08780594170093536,
|
| 200 |
+
"eval_runtime": 86.4118,
|
| 201 |
+
"eval_samples_per_second": 4.016,
|
| 202 |
+
"eval_steps_per_second": 0.671,
|
| 203 |
"step": 200
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 3.25,
|
| 207 |
+
"eval_accuracy": 1.0,
|
| 208 |
+
"eval_loss": 0.08780594170093536,
|
| 209 |
+
"eval_runtime": 76.2242,
|
| 210 |
+
"eval_samples_per_second": 4.552,
|
| 211 |
+
"eval_steps_per_second": 0.761,
|
| 212 |
"step": 200
|
| 213 |
}
|
| 214 |
],
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5048
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0976022ff1e3e4b6c6ed3d4b9f4fbf236be50b91b6fc1edad006409b3e04fd11
|
| 3 |
size 5048
|