Model save

README.md

@@ -4,8 +4,6 @@ license: cc-by-nc-4.0
 base_model: MCG-NJU/videomae-base
 tags:
 - generated_from_trainer
-metrics:
-- accuracy
 model-index:
 - name: videomae-base-finetuned-T3-subset
   results: []
@@ -17,9 +15,6 @@ should probably proofread and complete it, then remove this comment. -->
 # videomae-base-finetuned-T3-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: 2.0573
-- Accuracy: 0.0
 
 ## Model description
 
@@ -42,8 +37,6 @@ The following hyperparameters were used during training:
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 16
 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
@@ -53,8 +46,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| No log        | 1.0   | 1    | 2.0573          | 0.0      |
-| No log        | 2.0   | 2    | 2.0573          | 0.0      |
+| No log        | 1.0   | 2    | 2.2516          | 0.0      |
 
 
 ### Framework versions

checkpoints/README.md

@@ -0,0 +1,65 @@
+---
+library_name: transformers
+license: cc-by-nc-4.0
+base_model: MCG-NJU/videomae-base
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: checkpoints
+  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. -->
+
+# checkpoints
+
+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: 2.1221
+- Accuracy: 0.0
+
+## 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: 8
+- eval_batch_size: 8
+- seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 32
+- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.1
+- training_steps: 2
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| No log        | 1.0   | 1    | 2.1221          | 0.0      |
+| No log        | 2.0   | 2    | 2.1221          | 0.0      |
+
+
+### Framework versions
+
+- Transformers 4.48.0
+- Pytorch 2.1.2+cu118
+- Datasets 3.2.0
+- Tokenizers 0.21.0

checkpoints/checkpoint-1/config.json

@@ -0,0 +1,51 @@
+{
+  "_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": "Fasten windlass",
+    "1": "Identify wound",
+    "2": "Position tourniquet",
+    "3": "Take tourniquet",
+    "4": "Take windlass",
+    "5": "Tighten strap",
+    "6": "Twist windlass",
+    "7": "Verify hemostasis"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "Fasten windlass": 0,
+    "Identify wound": 1,
+    "Position tourniquet": 2,
+    "Take tourniquet": 3,
+    "Take windlass": 4,
+    "Tighten strap": 5,
+    "Twist windlass": 6,
+    "Verify hemostasis": 7
+  },
+  "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": "4.48.0",
+  "tubelet_size": 2,
+  "use_mean_pooling": false
+}

checkpoints/checkpoint-1/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5aa68c9516ee2dfbb6bd58294f24184c261236f7294dd860e2cab29b0d0353f
+size 344955832

checkpoints/checkpoint-1/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32d0975d4591a2f71b8ccdc2da022ccf1ffd063a480179138c68fc35bdf64e4f
+size 690022458

checkpoints/checkpoint-1/preprocessor_config.json

@@ -0,0 +1,26 @@
+{
+  "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
+  }
+}

checkpoints/checkpoint-1/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ac5c921b80607bc6b542eec40c2324594db2b30c39cd6ead0c5c2d4ac8e298e
+size 14244

checkpoints/checkpoint-1/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccba1bf67e8aaf8487a248eb4b14c07b2ea7e401afbce1dfa2f005cd57fb4b68
+size 1064

checkpoints/checkpoint-1/trainer_state.json

@@ -0,0 +1,42 @@
+{
+  "best_metric": 0.0,
+  "best_model_checkpoint": "./videomae-base-finetuned-T3-subset/checkpoints\\checkpoint-1",
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 1,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.0,
+      "eval_loss": 2.0771484375,
+      "eval_runtime": 9.991,
+      "eval_samples_per_second": 0.3,
+      "eval_steps_per_second": 0.1,
+      "step": 1
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.4953575179354112e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/checkpoint-1/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60f7d91193f890b148baf05c2bb2458233093c5e6eab7cb0a5030801b39e1667
+size 5368

checkpoints/checkpoint-2/config.json

@@ -0,0 +1,51 @@
+{
+  "_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": "Fasten windlass",
+    "1": "Identify wound",
+    "2": "Position tourniquet",
+    "3": "Take tourniquet",
+    "4": "Take windlass",
+    "5": "Tighten strap",
+    "6": "Twist windlass",
+    "7": "Verify hemostasis"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "Fasten windlass": 0,
+    "Identify wound": 1,
+    "Position tourniquet": 2,
+    "Take tourniquet": 3,
+    "Take windlass": 4,
+    "Tighten strap": 5,
+    "Twist windlass": 6,
+    "Verify hemostasis": 7
+  },
+  "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": "4.48.0",
+  "tubelet_size": 2,
+  "use_mean_pooling": false
+}

checkpoints/checkpoint-2/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5aa68c9516ee2dfbb6bd58294f24184c261236f7294dd860e2cab29b0d0353f
+size 344955832

checkpoints/checkpoint-2/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32d0975d4591a2f71b8ccdc2da022ccf1ffd063a480179138c68fc35bdf64e4f
+size 690022458

checkpoints/checkpoint-2/preprocessor_config.json

@@ -0,0 +1,26 @@
+{
+  "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
+  }
+}

checkpoints/checkpoint-2/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5544616f30ebd3d2302ffada375b6923cf93fab8ed37fd6d397c5fa01d34c2d5
+size 14244

checkpoints/checkpoint-2/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ccba1bf67e8aaf8487a248eb4b14c07b2ea7e401afbce1dfa2f005cd57fb4b68
+size 1064

checkpoints/checkpoint-2/trainer_state.json

@@ -0,0 +1,51 @@
+{
+  "best_metric": 0.0,
+  "best_model_checkpoint": "./videomae-base-finetuned-T3-subset/checkpoints\\checkpoint-1",
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 2,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.0,
+      "eval_loss": 2.0771484375,
+      "eval_runtime": 9.991,
+      "eval_samples_per_second": 0.3,
+      "eval_steps_per_second": 0.1,
+      "step": 1
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.0,
+      "eval_loss": 2.0771484375,
+      "eval_runtime": 10.1305,
+      "eval_samples_per_second": 0.296,
+      "eval_steps_per_second": 0.099,
+      "step": 2
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "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": 2.9907150358708224e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/checkpoint-2/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60f7d91193f890b148baf05c2bb2458233093c5e6eab7cb0a5030801b39e1667
+size 5368

checkpoints/config.json

@@ -0,0 +1,51 @@
+{
+  "_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": "Fasten windlass",
+    "1": "Identify wound",
+    "2": "Position tourniquet",
+    "3": "Take tourniquet",
+    "4": "Take windlass",
+    "5": "Tighten strap",
+    "6": "Twist windlass",
+    "7": "Verify hemostasis"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "Fasten windlass": 0,
+    "Identify wound": 1,
+    "Position tourniquet": 2,
+    "Take tourniquet": 3,
+    "Take windlass": 4,
+    "Tighten strap": 5,
+    "Twist windlass": 6,
+    "Verify hemostasis": 7
+  },
+  "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": "4.48.0",
+  "tubelet_size": 2,
+  "use_mean_pooling": false
+}

checkpoints/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5aa68c9516ee2dfbb6bd58294f24184c261236f7294dd860e2cab29b0d0353f
+size 344955832

checkpoints/preprocessor_config.json

@@ -0,0 +1,26 @@
+{
+  "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
+  }
+}

checkpoints/trainer_state.json

@@ -0,0 +1,60 @@
+{
+  "best_metric": 0.0,
+  "best_model_checkpoint": "./videomae-base-finetuned-T3-subset/checkpoints\\checkpoint-1",
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 2,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.0,
+      "eval_loss": 2.0771484375,
+      "eval_runtime": 9.991,
+      "eval_samples_per_second": 0.3,
+      "eval_steps_per_second": 0.1,
+      "step": 1
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.0,
+      "eval_loss": 2.0771484375,
+      "eval_runtime": 10.1305,
+      "eval_samples_per_second": 0.296,
+      "eval_steps_per_second": 0.099,
+      "step": 2
+    },
+    {
+      "epoch": 2.0,
+      "step": 2,
+      "total_flos": 2.9907150358708224e+16,
+      "train_loss": 1.072479248046875,
+      "train_runtime": 144.4028,
+      "train_samples_per_second": 0.443,
+      "train_steps_per_second": 0.014
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "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": 2.9907150358708224e+16,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoints/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60f7d91193f890b148baf05c2bb2458233093c5e6eab7cb0a5030801b39e1667
+size 5368

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:808fd46607d8d60f7b77b6afc861ae06572cbb7418f643b20bca12891e7dea41
+oid sha256:7134cb35be2820eb758d526c7422cb31311f6ebf3499d1120369e441b18a377c
 size 344955832

runs/Jan15_04-19-36_MININT-05KILTG/events.out.tfevents.1736911177.MININT-05KILTG.5092.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2f94f1aee8734787bad52bbaeaf800931be189b8d6e9db63a2c86cb15645bc6e
+size 6356

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.0,
+  "best_metric": 0.3333333333333333,
   "best_model_checkpoint": "videomae-base-finetuned-T3-subset\\checkpoint-1",
   "epoch": 2.0,
   "eval_steps": 500,
@@ -10,29 +10,29 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.0,
-      "eval_loss": 2.0572917461395264,
-      "eval_runtime": 13.4691,
-      "eval_samples_per_second": 0.223,
-      "eval_steps_per_second": 0.074,
+      "eval_accuracy": 0.3333333333333333,
+      "eval_loss": 1.8824869394302368,
+      "eval_runtime": 23.5572,
+      "eval_samples_per_second": 0.127,
+      "eval_steps_per_second": 0.042,
       "step": 1
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.0,
-      "eval_loss": 2.0572917461395264,
-      "eval_runtime": 12.9308,
-      "eval_samples_per_second": 0.232,
-      "eval_steps_per_second": 0.077,
+      "eval_accuracy": 0.3333333333333333,
+      "eval_loss": 2.037109375,
+      "eval_runtime": 26.8131,
+      "eval_samples_per_second": 0.112,
+      "eval_steps_per_second": 0.037,
       "step": 2
     },
     {
       "epoch": 2.0,
       "step": 2,
       "total_flos": 2.9907150358708224e+16,
-      "train_loss": 2.170654296875,
-      "train_runtime": 149.1885,
-      "train_samples_per_second": 0.214,
+      "train_loss": 1.056182861328125,
+      "train_runtime": 157.6397,
+      "train_samples_per_second": 0.406,
       "train_steps_per_second": 0.013
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:479f106c7b076bce4bf7cded40dcd43e7523c442677ebe0f5605e3d6db9b3d15
+oid sha256:14a1dd1ccea6f90396f08776c4ea5bf6964f0219d32a2d57f2ee93b779fd1baa
 size 5368