rotated_maps

README.md

@@ -0,0 +1,80 @@
+---
+library_name: transformers
+license: apache-2.0
+base_model: google/vit-base-patch16-224-in21k
+tags:
+- image-classification
+- generated_from_trainer
+datasets:
+- imagefolder
+metrics:
+- accuracy
+model-index:
+- name: vit-base-patch16-224-in21k-rotated-dungeons-v12
+  results:
+  - task:
+      name: Image Classification
+      type: image-classification
+    dataset:
+      name: rotated_maps
+      type: imagefolder
+      config: default
+      split: validation
+      args: default
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.6666666666666666
+---
+
+<!-- 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. -->
+
+# vit-base-patch16-224-in21k-rotated-dungeons-v12
+
+This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the rotated_maps dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.4316
+- Accuracy: 0.6667
+
+## 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: 1e-05
+- train_batch_size: 4
+- eval_batch_size: 8
+- seed: 1024
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 27
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch   | Step | Validation Loss | Accuracy |
+|:-------------:|:-------:|:----:|:---------------:|:--------:|
+| 1.0811        | 8.3333  | 100  | 1.6611          | 0.4167   |
+| 0.6509        | 16.6667 | 200  | 1.4987          | 0.5      |
+| 0.5656        | 25.0    | 300  | 1.4316          | 0.6667   |
+
+
+### Framework versions
+
+- Transformers 4.44.2
+- Pytorch 2.5.0+cu121
+- Datasets 3.1.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 27.0,
+    "eval_accuracy": 0.6666666666666666,
+    "eval_loss": 1.4315592050552368,
+    "eval_runtime": 0.1454,
+    "eval_samples_per_second": 82.532,
+    "eval_steps_per_second": 13.755,
+    "total_flos": 9.624936417107558e+16,
+    "train_loss": 0.9185323832947531,
+    "train_runtime": 97.1154,
+    "train_samples_per_second": 12.789,
+    "train_steps_per_second": 3.336
+}

config.json

@@ -0,0 +1,42 @@
+{
+  "_name_or_path": "google/vit-base-patch16-224-in21k",
+  "architectures": [
+    "ViTForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "encoder_stride": 16,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "five",
+    "1": "four",
+    "2": "one",
+    "3": "three",
+    "4": "twelve",
+    "5": "two",
+    "6": "zero"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "five": "0",
+    "four": "1",
+    "one": "2",
+    "three": "3",
+    "twelve": "4",
+    "two": "5",
+    "zero": "6"
+  },
+  "layer_norm_eps": 1e-12,
+  "model_type": "vit",
+  "num_attention_heads": 12,
+  "num_channels": 3,
+  "num_hidden_layers": 12,
+  "patch_size": 16,
+  "problem_type": "single_label_classification",
+  "qkv_bias": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.44.2"
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 27.0,
+    "eval_accuracy": 0.6666666666666666,
+    "eval_loss": 1.4315592050552368,
+    "eval_runtime": 0.1454,
+    "eval_samples_per_second": 82.532,
+    "eval_steps_per_second": 13.755
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1893e4e1fd3068d520917356318f4cbb59b17f3dc378be67fad611a2bc2308c7
+size 343239356

preprocessor_config.json

@@ -0,0 +1,22 @@
+{
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "image_processor_type": "ViTImageProcessor",
+  "image_std": [
+    0.5,
+    0.5,
+    0.5
+  ],
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 224,
+    "width": 224
+  }
+}

runs/Nov02_20-31-19_be037f56c5ae/events.out.tfevents.1730579482.be037f56c5ae.1722.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:275107fb8a22e59f3fa3a8b9da121ea3bfff1c5daae51520dde5865b25a65f8d
+size 13174

runs/Nov02_20-31-19_be037f56c5ae/events.out.tfevents.1730579581.be037f56c5ae.1722.8

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:641d9a34997851f22471b26377acfb68b6ea542c8bed4e9a4b023276794e8914
+size 411

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 27.0,
+    "total_flos": 9.624936417107558e+16,
+    "train_loss": 0.9185323832947531,
+    "train_runtime": 97.1154,
+    "train_samples_per_second": 12.789,
+    "train_steps_per_second": 3.336
+}

trainer_state.json

@@ -0,0 +1,293 @@
+{
+  "best_metric": 1.4315592050552368,
+  "best_model_checkpoint": "vit-base-patch16-224-in21k-rotated-dungeons-v12/checkpoint-300",
+  "epoch": 27.0,
+  "eval_steps": 100,
+  "global_step": 324,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 3.6018190383911133,
+      "learning_rate": 9.691358024691358e-06,
+      "loss": 1.8918,
+      "step": 10
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 3.4979114532470703,
+      "learning_rate": 9.382716049382717e-06,
+      "loss": 1.7508,
+      "step": 20
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 4.02228307723999,
+      "learning_rate": 9.074074074074075e-06,
+      "loss": 1.5583,
+      "step": 30
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 3.0986037254333496,
+      "learning_rate": 8.765432098765432e-06,
+      "loss": 1.4717,
+      "step": 40
+    },
+    {
+      "epoch": 4.166666666666667,
+      "grad_norm": 3.573038339614868,
+      "learning_rate": 8.456790123456791e-06,
+      "loss": 1.4678,
+      "step": 50
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 4.513486385345459,
+      "learning_rate": 8.148148148148148e-06,
+      "loss": 1.3504,
+      "step": 60
+    },
+    {
+      "epoch": 5.833333333333333,
+      "grad_norm": 2.933511257171631,
+      "learning_rate": 7.839506172839507e-06,
+      "loss": 1.28,
+      "step": 70
+    },
+    {
+      "epoch": 6.666666666666667,
+      "grad_norm": 4.242460250854492,
+      "learning_rate": 7.530864197530865e-06,
+      "loss": 1.2036,
+      "step": 80
+    },
+    {
+      "epoch": 7.5,
+      "grad_norm": 4.462291240692139,
+      "learning_rate": 7.222222222222223e-06,
+      "loss": 1.1049,
+      "step": 90
+    },
+    {
+      "epoch": 8.333333333333334,
+      "grad_norm": 3.108102560043335,
+      "learning_rate": 6.913580246913581e-06,
+      "loss": 1.0811,
+      "step": 100
+    },
+    {
+      "epoch": 8.333333333333334,
+      "eval_accuracy": 0.4166666666666667,
+      "eval_loss": 1.6610513925552368,
+      "eval_runtime": 0.3767,
+      "eval_samples_per_second": 31.86,
+      "eval_steps_per_second": 5.31,
+      "step": 100
+    },
+    {
+      "epoch": 9.166666666666666,
+      "grad_norm": 2.9008240699768066,
+      "learning_rate": 6.60493827160494e-06,
+      "loss": 1.0747,
+      "step": 110
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 3.7455575466156006,
+      "learning_rate": 6.296296296296297e-06,
+      "loss": 0.941,
+      "step": 120
+    },
+    {
+      "epoch": 10.833333333333334,
+      "grad_norm": 2.9407384395599365,
+      "learning_rate": 5.9876543209876546e-06,
+      "loss": 0.8968,
+      "step": 130
+    },
+    {
+      "epoch": 11.666666666666666,
+      "grad_norm": 2.319361448287964,
+      "learning_rate": 5.6790123456790125e-06,
+      "loss": 0.8519,
+      "step": 140
+    },
+    {
+      "epoch": 12.5,
+      "grad_norm": 4.124772071838379,
+      "learning_rate": 5.370370370370371e-06,
+      "loss": 0.9155,
+      "step": 150
+    },
+    {
+      "epoch": 13.333333333333334,
+      "grad_norm": 2.6311306953430176,
+      "learning_rate": 5.061728395061729e-06,
+      "loss": 0.7229,
+      "step": 160
+    },
+    {
+      "epoch": 14.166666666666666,
+      "grad_norm": 2.6873831748962402,
+      "learning_rate": 4.753086419753087e-06,
+      "loss": 0.7839,
+      "step": 170
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 4.260179042816162,
+      "learning_rate": 4.444444444444444e-06,
+      "loss": 0.7357,
+      "step": 180
+    },
+    {
+      "epoch": 15.833333333333334,
+      "grad_norm": 2.457617998123169,
+      "learning_rate": 4.135802469135803e-06,
+      "loss": 0.7007,
+      "step": 190
+    },
+    {
+      "epoch": 16.666666666666668,
+      "grad_norm": 2.1463468074798584,
+      "learning_rate": 3.827160493827161e-06,
+      "loss": 0.6509,
+      "step": 200
+    },
+    {
+      "epoch": 16.666666666666668,
+      "eval_accuracy": 0.5,
+      "eval_loss": 1.4986978769302368,
+      "eval_runtime": 0.117,
+      "eval_samples_per_second": 102.54,
+      "eval_steps_per_second": 17.09,
+      "step": 200
+    },
+    {
+      "epoch": 17.5,
+      "grad_norm": 2.4582600593566895,
+      "learning_rate": 3.5185185185185187e-06,
+      "loss": 0.6903,
+      "step": 210
+    },
+    {
+      "epoch": 18.333333333333332,
+      "grad_norm": 1.9703677892684937,
+      "learning_rate": 3.2098765432098767e-06,
+      "loss": 0.656,
+      "step": 220
+    },
+    {
+      "epoch": 19.166666666666668,
+      "grad_norm": 2.4972715377807617,
+      "learning_rate": 2.901234567901235e-06,
+      "loss": 0.6353,
+      "step": 230
+    },
+    {
+      "epoch": 20.0,
+      "grad_norm": 2.1885712146759033,
+      "learning_rate": 2.5925925925925925e-06,
+      "loss": 0.6088,
+      "step": 240
+    },
+    {
+      "epoch": 20.833333333333332,
+      "grad_norm": 2.897416591644287,
+      "learning_rate": 2.283950617283951e-06,
+      "loss": 0.6143,
+      "step": 250
+    },
+    {
+      "epoch": 21.666666666666668,
+      "grad_norm": 1.803607702255249,
+      "learning_rate": 1.9753086419753087e-06,
+      "loss": 0.551,
+      "step": 260
+    },
+    {
+      "epoch": 22.5,
+      "grad_norm": 2.254380464553833,
+      "learning_rate": 1.6666666666666667e-06,
+      "loss": 0.5492,
+      "step": 270
+    },
+    {
+      "epoch": 23.333333333333332,
+      "grad_norm": 1.718522548675537,
+      "learning_rate": 1.3580246913580248e-06,
+      "loss": 0.5485,
+      "step": 280
+    },
+    {
+      "epoch": 24.166666666666668,
+      "grad_norm": 2.0524728298187256,
+      "learning_rate": 1.0493827160493827e-06,
+      "loss": 0.5897,
+      "step": 290
+    },
+    {
+      "epoch": 25.0,
+      "grad_norm": 3.3403310775756836,
+      "learning_rate": 7.407407407407407e-07,
+      "loss": 0.5656,
+      "step": 300
+    },
+    {
+      "epoch": 25.0,
+      "eval_accuracy": 0.6666666666666666,
+      "eval_loss": 1.4315592050552368,
+      "eval_runtime": 0.138,
+      "eval_samples_per_second": 86.966,
+      "eval_steps_per_second": 14.494,
+      "step": 300
+    },
+    {
+      "epoch": 25.833333333333332,
+      "grad_norm": 2.1986382007598877,
+      "learning_rate": 4.320987654320988e-07,
+      "loss": 0.5497,
+      "step": 310
+    },
+    {
+      "epoch": 26.666666666666668,
+      "grad_norm": 1.6566905975341797,
+      "learning_rate": 1.234567901234568e-07,
+      "loss": 0.5395,
+      "step": 320
+    },
+    {
+      "epoch": 27.0,
+      "step": 324,
+      "total_flos": 9.624936417107558e+16,
+      "train_loss": 0.9185323832947531,
+      "train_runtime": 97.1154,
+      "train_samples_per_second": 12.789,
+      "train_steps_per_second": 3.336
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 324,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 27,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 9.624936417107558e+16,
+  "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 sha256:281636cbda2469109803566c0d502f19c1cb5cc7c427a8da9387ee4acc6645a2
+size 5240