End of training

README.md

@@ -0,0 +1,64 @@
+---
+license: cc-by-nc-4.0
+base_model: MCG-NJU/videomae-base
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: videomae-base-finetuned-ucf101-subset
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# videomae-base-finetuned-ucf101-subset
+
+This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3131
+- Accuracy: 0.9161
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 4
+- eval_batch_size: 4
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- training_steps: 300
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 1.6397        | 0.25  | 75   | 1.5641          | 0.4      |
+| 0.8215        | 1.25  | 150  | 0.8054          | 0.6429   |
+| 0.2651        | 2.25  | 225  | 0.4286          | 0.8429   |
+| 0.202         | 3.25  | 300  | 0.3048          | 0.9      |
+
+
+### Framework versions
+
+- Transformers 4.41.2
+- Pytorch 2.3.1+cu121
+- Datasets 2.20.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.25,
+    "eval_accuracy": 0.9161290322580645,
+    "eval_loss": 0.3130556344985962,
+    "eval_runtime": 44.5775,
+    "eval_samples_per_second": 3.477,
+    "eval_steps_per_second": 0.875
+}

config.json

@@ -0,0 +1,55 @@
+{
+  "_name_or_path": "MCG-NJU/videomae-base",
+  "architectures": [
+    "VideoMAEForVideoClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "decoder_hidden_size": 384,
+  "decoder_intermediate_size": 1536,
+  "decoder_num_attention_heads": 6,
+  "decoder_num_hidden_layers": 4,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "ApplyEyeMakeup",
+    "1": "ApplyLipstick",
+    "2": "Archery",
+    "3": "BabyCrawling",
+    "4": "BalanceBeam",
+    "5": "BandMarching",
+    "6": "BaseballPitch",
+    "7": "Basketball",
+    "8": "BasketballDunk",
+    "9": "BenchPress"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "ApplyEyeMakeup": 0,
+    "ApplyLipstick": 1,
+    "Archery": 2,
+    "BabyCrawling": 3,
+    "BalanceBeam": 4,
+    "BandMarching": 5,
+    "BaseballPitch": 6,
+    "Basketball": 7,
+    "BasketballDunk": 8,
+    "BenchPress": 9
+  },
+  "layer_norm_eps": 1e-12,
+  "model_type": "videomae",
+  "norm_pix_loss": true,
+  "num_attention_heads": 12,
+  "num_channels": 3,
+  "num_frames": 16,
+  "num_hidden_layers": 12,
+  "patch_size": 16,
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.41.2",
+  "tubelet_size": 2,
+  "use_mean_pooling": false
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d6e5f552194670320217649faf08e0059de358c41b5160a46b3c9ae254b76b4
+size 344961984

preprocessor_config.json

@@ -0,0 +1,42 @@
+{
+  "_valid_processor_keys": [
+    "videos",
+    "do_resize",
+    "size",
+    "resample",
+    "do_center_crop",
+    "crop_size",
+    "do_rescale",
+    "rescale_factor",
+    "do_normalize",
+    "image_mean",
+    "image_std",
+    "return_tensors",
+    "data_format",
+    "input_data_format"
+  ],
+  "crop_size": {
+    "height": 224,
+    "width": 224
+  },
+  "do_center_crop": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_processor_type": "VideoMAEImageProcessor",
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "shortest_edge": 224
+  }
+}

runs/Jun22_22-48-22_35076ae28850/events.out.tfevents.1719096505.35076ae28850.3689.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8e8f031b9953ddf144ed9e7e133b143ae168578359e6d8e6d6e3712dcb85612
+size 16348

runs/Jun22_22-50-44_35076ae28850/events.out.tfevents.1719096665.35076ae28850.3689.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19f35a29adb96813698ef5020dc762c44d4c843bc84c0ae24ef030c74ef1559e
+size 5507

runs/Jun22_22-51-19_35076ae28850/events.out.tfevents.1719096690.35076ae28850.3689.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55b514b99afc12c1cdfc806cb0dbfd3984496c3d5782e4539dbc481798126981
+size 13429

runs/Jun22_22-51-19_35076ae28850/events.out.tfevents.1719100722.35076ae28850.3689.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06ddf6c60190867ab75076beef5c938d1909383400cdb72b9bf9f9fe2b82f456
+size 734

test_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 3.25,
+    "eval_accuracy": 0.9161290322580645,
+    "eval_loss": 0.3130556344985962,
+    "eval_runtime": 44.5775,
+    "eval_samples_per_second": 3.477,
+    "eval_steps_per_second": 0.875
+}

trainer_state.json

@@ -0,0 +1,306 @@
+{
+  "best_metric": 0.9,
+  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-300",
+  "epoch": 3.25,
+  "eval_steps": 500,
+  "global_step": 300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.03333333333333333,
+      "grad_norm": 20.201610565185547,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 2.444,
+      "step": 10
+    },
+    {
+      "epoch": 0.06666666666666667,
+      "grad_norm": 12.654314994812012,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 2.3019,
+      "step": 20
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 12.947044372558594,
+      "learning_rate": 5e-05,
+      "loss": 2.2352,
+      "step": 30
+    },
+    {
+      "epoch": 0.13333333333333333,
+      "grad_norm": 11.306678771972656,
+      "learning_rate": 4.814814814814815e-05,
+      "loss": 2.1503,
+      "step": 40
+    },
+    {
+      "epoch": 0.16666666666666666,
+      "grad_norm": 11.73679256439209,
+      "learning_rate": 4.62962962962963e-05,
+      "loss": 2.0142,
+      "step": 50
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 17.422954559326172,
+      "learning_rate": 4.4444444444444447e-05,
+      "loss": 1.9569,
+      "step": 60
+    },
+    {
+      "epoch": 0.23333333333333334,
+      "grad_norm": 12.51447868347168,
+      "learning_rate": 4.259259259259259e-05,
+      "loss": 1.6397,
+      "step": 70
+    },
+    {
+      "epoch": 0.25,
+      "eval_accuracy": 0.4,
+      "eval_loss": 1.5641058683395386,
+      "eval_runtime": 19.6018,
+      "eval_samples_per_second": 3.571,
+      "eval_steps_per_second": 0.918,
+      "step": 75
+    },
+    {
+      "epoch": 1.0166666666666666,
+      "grad_norm": 13.403763771057129,
+      "learning_rate": 4.074074074074074e-05,
+      "loss": 1.3891,
+      "step": 80
+    },
+    {
+      "epoch": 1.05,
+      "grad_norm": 11.36450481414795,
+      "learning_rate": 3.888888888888889e-05,
+      "loss": 0.9598,
+      "step": 90
+    },
+    {
+      "epoch": 1.0833333333333333,
+      "grad_norm": 14.846416473388672,
+      "learning_rate": 3.7037037037037037e-05,
+      "loss": 0.8928,
+      "step": 100
+    },
+    {
+      "epoch": 1.1166666666666667,
+      "grad_norm": 9.266874313354492,
+      "learning_rate": 3.518518518518519e-05,
+      "loss": 0.8165,
+      "step": 110
+    },
+    {
+      "epoch": 1.15,
+      "grad_norm": 21.095354080200195,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.8177,
+      "step": 120
+    },
+    {
+      "epoch": 1.1833333333333333,
+      "grad_norm": 2.8057732582092285,
+      "learning_rate": 3.148148148148148e-05,
+      "loss": 0.6388,
+      "step": 130
+    },
+    {
+      "epoch": 1.2166666666666668,
+      "grad_norm": 7.6457672119140625,
+      "learning_rate": 2.962962962962963e-05,
+      "loss": 0.6413,
+      "step": 140
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 15.42055892944336,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.8215,
+      "step": 150
+    },
+    {
+      "epoch": 1.25,
+      "eval_accuracy": 0.6428571428571429,
+      "eval_loss": 0.8053566813468933,
+      "eval_runtime": 20.071,
+      "eval_samples_per_second": 3.488,
+      "eval_steps_per_second": 0.897,
+      "step": 150
+    },
+    {
+      "epoch": 2.033333333333333,
+      "grad_norm": 30.937313079833984,
+      "learning_rate": 2.5925925925925925e-05,
+      "loss": 0.6518,
+      "step": 160
+    },
+    {
+      "epoch": 2.066666666666667,
+      "grad_norm": 3.2372474670410156,
+      "learning_rate": 2.4074074074074074e-05,
+      "loss": 0.5538,
+      "step": 170
+    },
+    {
+      "epoch": 2.1,
+      "grad_norm": 6.289671421051025,
+      "learning_rate": 2.2222222222222223e-05,
+      "loss": 0.3373,
+      "step": 180
+    },
+    {
+      "epoch": 2.1333333333333333,
+      "grad_norm": 28.10116195678711,
+      "learning_rate": 2.037037037037037e-05,
+      "loss": 0.4068,
+      "step": 190
+    },
+    {
+      "epoch": 2.1666666666666665,
+      "grad_norm": 2.922621726989746,
+      "learning_rate": 1.8518518518518518e-05,
+      "loss": 0.3649,
+      "step": 200
+    },
+    {
+      "epoch": 2.2,
+      "grad_norm": 1.8260467052459717,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.4159,
+      "step": 210
+    },
+    {
+      "epoch": 2.2333333333333334,
+      "grad_norm": 7.952264308929443,
+      "learning_rate": 1.4814814814814815e-05,
+      "loss": 0.2651,
+      "step": 220
+    },
+    {
+      "epoch": 2.25,
+      "eval_accuracy": 0.8428571428571429,
+      "eval_loss": 0.4285981357097626,
+      "eval_runtime": 19.6403,
+      "eval_samples_per_second": 3.564,
+      "eval_steps_per_second": 0.916,
+      "step": 225
+    },
+    {
+      "epoch": 3.0166666666666666,
+      "grad_norm": 8.908254623413086,
+      "learning_rate": 1.2962962962962962e-05,
+      "loss": 0.3064,
+      "step": 230
+    },
+    {
+      "epoch": 3.05,
+      "grad_norm": 27.685514450073242,
+      "learning_rate": 1.1111111111111112e-05,
+      "loss": 0.2469,
+      "step": 240
+    },
+    {
+      "epoch": 3.0833333333333335,
+      "grad_norm": 6.450607776641846,
+      "learning_rate": 9.259259259259259e-06,
+      "loss": 0.31,
+      "step": 250
+    },
+    {
+      "epoch": 3.1166666666666667,
+      "grad_norm": 0.25113508105278015,
+      "learning_rate": 7.4074074074074075e-06,
+      "loss": 0.1282,
+      "step": 260
+    },
+    {
+      "epoch": 3.15,
+      "grad_norm": 4.858123779296875,
+      "learning_rate": 5.555555555555556e-06,
+      "loss": 0.1547,
+      "step": 270
+    },
+    {
+      "epoch": 3.183333333333333,
+      "grad_norm": 10.586645126342773,
+      "learning_rate": 3.7037037037037037e-06,
+      "loss": 0.1706,
+      "step": 280
+    },
+    {
+      "epoch": 3.216666666666667,
+      "grad_norm": 0.892095148563385,
+      "learning_rate": 1.8518518518518519e-06,
+      "loss": 0.2138,
+      "step": 290
+    },
+    {
+      "epoch": 3.25,
+      "grad_norm": 4.021522521972656,
+      "learning_rate": 0.0,
+      "loss": 0.202,
+      "step": 300
+    },
+    {
+      "epoch": 3.25,
+      "eval_accuracy": 0.9,
+      "eval_loss": 0.3048226535320282,
+      "eval_runtime": 19.4139,
+      "eval_samples_per_second": 3.606,
+      "eval_steps_per_second": 0.927,
+      "step": 300
+    },
+    {
+      "epoch": 3.25,
+      "step": 300,
+      "total_flos": 1.495384188125184e+18,
+      "train_loss": 0.8815905849138895,
+      "train_runtime": 827.7558,
+      "train_samples_per_second": 1.45,
+      "train_steps_per_second": 0.362
+    },
+    {
+      "epoch": 3.25,
+      "eval_accuracy": 0.9166666666666666,
+      "eval_loss": 0.31168052554130554,
+      "eval_runtime": 54.4422,
+      "eval_samples_per_second": 2.865,
+      "eval_steps_per_second": 0.716,
+      "step": 300
+    },
+    {
+      "epoch": 3.25,
+      "eval_accuracy": 0.9161290322580645,
+      "eval_loss": 0.3130556344985962,
+      "eval_runtime": 44.5775,
+      "eval_samples_per_second": 3.477,
+      "eval_steps_per_second": 0.875,
+      "step": 300
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 300,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 9223372036854775807,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.495384188125184e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a8a5b8456ba633c6bb1cab32266ed6c56d30e7211b126d1b0740b0bbff31fb4
+size 5176