initialcommit

README.md

@@ -0,0 +1,65 @@
+---
+license: apache-2.0
+base_model: google/vit-base-patch16-224
+tags:
+- image-classification
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: vit-base-pets
+  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. -->
+
+# vit-base-pets
+
+This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the pcuenq/oxford-pets dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0037
+- Accuracy: 0.9995
+
+## 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: 0.0003
+- train_batch_size: 64
+- eval_batch_size: 8
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- num_epochs: 4
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 0.3546        | 0.86  | 100  | 0.1640          | 0.9545   |
+| 0.0922        | 1.72  | 200  | 0.0596          | 0.9850   |
+| 0.0421        | 2.59  | 300  | 0.0214          | 0.9939   |
+| 0.0087        | 3.45  | 400  | 0.0037          | 0.9995   |
+
+
+### Framework versions
+
+- Transformers 4.39.2
+- Pytorch 2.2.1+cu121
+- Datasets 2.18.0
+- Tokenizers 0.15.2

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.0,
+    "total_flos": 2.2913817801515827e+18,
+    "train_loss": 0.18540822506223903,
+    "train_runtime": 740.8045,
+    "train_samples_per_second": 39.903,
+    "train_steps_per_second": 0.626
+}

config.json

@@ -0,0 +1,102 @@
+{
+  "_name_or_path": "google/vit-base-patch16-224",
+  "architectures": [
+    "ViTForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "encoder_stride": 16,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "yorkshire terrier Dog",
+    "1": "Birman Cat",
+    "10": "pug Dog",
+    "11": "Persian Cat",
+    "12": "english setter Dog",
+    "13": "american pit bull terrier Dog",
+    "14": "chihuahua Dog",
+    "15": "pomeranian Dog",
+    "16": "great pyrenees Dog",
+    "17": "scottish terrier Dog",
+    "18": "miniature pinscher Dog",
+    "19": "Sphynx Cat",
+    "2": "german shorthaired Dog",
+    "20": "Maine Coon Cat",
+    "21": "saint bernard Dog",
+    "22": "Siamese Cat",
+    "23": "american bulldog Dog",
+    "24": "newfoundland Dog",
+    "25": "Abyssinian Cat",
+    "26": "leonberger Dog",
+    "27": "beagle Dog",
+    "28": "Egyptian Mau Cat",
+    "29": "boxer Dog",
+    "3": "keeshond Dog",
+    "30": "British Shorthair Cat",
+    "31": "staffordshire bull terrier Dog",
+    "32": "samoyed Dog",
+    "33": "basset hound Dog",
+    "34": "Bengal Cat",
+    "35": "Bombay Cat",
+    "36": "Russian Blue Cat",
+    "4": "havanese Dog",
+    "5": "wheaten terrier Dog",
+    "6": "shiba inu Dog",
+    "7": "Ragdoll Cat",
+    "8": "japanese chin Dog",
+    "9": "english cocker spaniel Dog"
+  },
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "Abyssinian Cat": "25",
+    "Bengal Cat": "34",
+    "Birman Cat": "1",
+    "Bombay Cat": "35",
+    "British Shorthair Cat": "30",
+    "Egyptian Mau Cat": "28",
+    "Maine Coon Cat": "20",
+    "Persian Cat": "11",
+    "Ragdoll Cat": "7",
+    "Russian Blue Cat": "36",
+    "Siamese Cat": "22",
+    "Sphynx Cat": "19",
+    "american bulldog Dog": "23",
+    "american pit bull terrier Dog": "13",
+    "basset hound Dog": "33",
+    "beagle Dog": "27",
+    "boxer Dog": "29",
+    "chihuahua Dog": "14",
+    "english cocker spaniel Dog": "9",
+    "english setter Dog": "12",
+    "german shorthaired Dog": "2",
+    "great pyrenees Dog": "16",
+    "havanese Dog": "4",
+    "japanese chin Dog": "8",
+    "keeshond Dog": "3",
+    "leonberger Dog": "26",
+    "miniature pinscher Dog": "18",
+    "newfoundland Dog": "24",
+    "pomeranian Dog": "15",
+    "pug Dog": "10",
+    "saint bernard Dog": "21",
+    "samoyed Dog": "32",
+    "scottish terrier Dog": "17",
+    "shiba inu Dog": "6",
+    "staffordshire bull terrier Dog": "31",
+    "wheaten terrier Dog": "5",
+    "yorkshire terrier Dog": "0"
+  },
+  "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": "4.39.2"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b2184cabd333fc0f02fa6c917e9aaec72c255443c28de6752429d9b05a440e5
+size 343331644

runs/Mar30_07-36-47_7ccd926a2ffc/events.out.tfevents.1711784216.7ccd926a2ffc.12602.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab5d001998ff2fa79116898263451ae1b1bbc4e05c6a7437422d7b9c43211b22
+size 17991

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 4.0,
+    "total_flos": 2.2913817801515827e+18,
+    "train_loss": 0.18540822506223903,
+    "train_runtime": 740.8045,
+    "train_samples_per_second": 39.903,
+    "train_steps_per_second": 0.626
+}

trainer_state.json

@@ -0,0 +1,388 @@
+{
+  "best_metric": 0.0037413176614791155,
+  "best_model_checkpoint": "./vit-base-pets/checkpoint-400",
+  "epoch": 4.0,
+  "eval_steps": 100,
+  "global_step": 464,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.09,
+      "grad_norm": 3.7365024089813232,
+      "learning_rate": 0.00029353448275862065,
+      "loss": 2.333,
+      "step": 10
+    },
+    {
+      "epoch": 0.17,
+      "grad_norm": 2.494818925857544,
+      "learning_rate": 0.0002870689655172413,
+      "loss": 0.7182,
+      "step": 20
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 2.9715936183929443,
+      "learning_rate": 0.00028060344827586205,
+      "loss": 0.4629,
+      "step": 30
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 3.1811089515686035,
+      "learning_rate": 0.0002741379310344827,
+      "loss": 0.4464,
+      "step": 40
+    },
+    {
+      "epoch": 0.43,
+      "grad_norm": 3.517517566680908,
+      "learning_rate": 0.00026767241379310345,
+      "loss": 0.3636,
+      "step": 50
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 1.850231647491455,
+      "learning_rate": 0.00026120689655172413,
+      "loss": 0.3916,
+      "step": 60
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 3.451078414916992,
+      "learning_rate": 0.0002547413793103448,
+      "loss": 0.423,
+      "step": 70
+    },
+    {
+      "epoch": 0.69,
+      "grad_norm": 3.729478359222412,
+      "learning_rate": 0.0002482758620689655,
+      "loss": 0.4177,
+      "step": 80
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 3.4009933471679688,
+      "learning_rate": 0.00024181034482758618,
+      "loss": 0.3164,
+      "step": 90
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 2.6017422676086426,
+      "learning_rate": 0.00023534482758620685,
+      "loss": 0.3546,
+      "step": 100
+    },
+    {
+      "epoch": 0.86,
+      "eval_accuracy": 0.9545331529093369,
+      "eval_loss": 0.16398410499095917,
+      "eval_runtime": 66.2449,
+      "eval_samples_per_second": 111.556,
+      "eval_steps_per_second": 13.948,
+      "step": 100
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 2.4210205078125,
+      "learning_rate": 0.00022887931034482758,
+      "loss": 0.3267,
+      "step": 110
+    },
+    {
+      "epoch": 1.03,
+      "grad_norm": 1.4406899213790894,
+      "learning_rate": 0.00022241379310344826,
+      "loss": 0.2172,
+      "step": 120
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 1.6676567792892456,
+      "learning_rate": 0.00021594827586206896,
+      "loss": 0.1671,
+      "step": 130
+    },
+    {
+      "epoch": 1.21,
+      "grad_norm": 1.7579220533370972,
+      "learning_rate": 0.00020948275862068963,
+      "loss": 0.1416,
+      "step": 140
+    },
+    {
+      "epoch": 1.29,
+      "grad_norm": 0.9925330281257629,
+      "learning_rate": 0.00020301724137931034,
+      "loss": 0.1289,
+      "step": 150
+    },
+    {
+      "epoch": 1.38,
+      "grad_norm": 1.1069083213806152,
+      "learning_rate": 0.000196551724137931,
+      "loss": 0.1199,
+      "step": 160
+    },
+    {
+      "epoch": 1.47,
+      "grad_norm": 2.214843273162842,
+      "learning_rate": 0.00019008620689655169,
+      "loss": 0.1204,
+      "step": 170
+    },
+    {
+      "epoch": 1.55,
+      "grad_norm": 2.182187080383301,
+      "learning_rate": 0.0001836206896551724,
+      "loss": 0.1155,
+      "step": 180
+    },
+    {
+      "epoch": 1.64,
+      "grad_norm": 2.003282308578491,
+      "learning_rate": 0.0001771551724137931,
+      "loss": 0.1319,
+      "step": 190
+    },
+    {
+      "epoch": 1.72,
+      "grad_norm": 2.0936930179595947,
+      "learning_rate": 0.0001706896551724138,
+      "loss": 0.0922,
+      "step": 200
+    },
+    {
+      "epoch": 1.72,
+      "eval_accuracy": 0.984979702300406,
+      "eval_loss": 0.05955846235156059,
+      "eval_runtime": 66.0945,
+      "eval_samples_per_second": 111.81,
+      "eval_steps_per_second": 13.98,
+      "step": 200
+    },
+    {
+      "epoch": 1.81,
+      "grad_norm": 1.8309838771820068,
+      "learning_rate": 0.00016422413793103446,
+      "loss": 0.1028,
+      "step": 210
+    },
+    {
+      "epoch": 1.9,
+      "grad_norm": 1.588274598121643,
+      "learning_rate": 0.00015775862068965517,
+      "loss": 0.1285,
+      "step": 220
+    },
+    {
+      "epoch": 1.98,
+      "grad_norm": 3.1869797706604004,
+      "learning_rate": 0.00015129310344827584,
+      "loss": 0.1091,
+      "step": 230
+    },
+    {
+      "epoch": 2.07,
+      "grad_norm": 0.7886011004447937,
+      "learning_rate": 0.00014482758620689654,
+      "loss": 0.0681,
+      "step": 240
+    },
+    {
+      "epoch": 2.16,
+      "grad_norm": 1.0175777673721313,
+      "learning_rate": 0.00013836206896551724,
+      "loss": 0.062,
+      "step": 250
+    },
+    {
+      "epoch": 2.24,
+      "grad_norm": 0.5733941197395325,
+      "learning_rate": 0.00013189655172413792,
+      "loss": 0.0447,
+      "step": 260
+    },
+    {
+      "epoch": 2.33,
+      "grad_norm": 0.6709829568862915,
+      "learning_rate": 0.0001254310344827586,
+      "loss": 0.0352,
+      "step": 270
+    },
+    {
+      "epoch": 2.41,
+      "grad_norm": 0.10771916061639786,
+      "learning_rate": 0.00011896551724137931,
+      "loss": 0.0281,
+      "step": 280
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.8952343463897705,
+      "learning_rate": 0.0001125,
+      "loss": 0.0401,
+      "step": 290
+    },
+    {
+      "epoch": 2.59,
+      "grad_norm": 0.05457449331879616,
+      "learning_rate": 0.00010603448275862067,
+      "loss": 0.0421,
+      "step": 300
+    },
+    {
+      "epoch": 2.59,
+      "eval_accuracy": 0.9939106901217862,
+      "eval_loss": 0.021390045061707497,
+      "eval_runtime": 66.2486,
+      "eval_samples_per_second": 111.55,
+      "eval_steps_per_second": 13.947,
+      "step": 300
+    },
+    {
+      "epoch": 2.67,
+      "grad_norm": 0.14218959212303162,
+      "learning_rate": 9.956896551724137e-05,
+      "loss": 0.0364,
+      "step": 310
+    },
+    {
+      "epoch": 2.76,
+      "grad_norm": 0.8931805491447449,
+      "learning_rate": 9.310344827586206e-05,
+      "loss": 0.0232,
+      "step": 320
+    },
+    {
+      "epoch": 2.84,
+      "grad_norm": 0.25261449813842773,
+      "learning_rate": 8.663793103448275e-05,
+      "loss": 0.0134,
+      "step": 330
+    },
+    {
+      "epoch": 2.93,
+      "grad_norm": 0.07521424442529678,
+      "learning_rate": 8.017241379310344e-05,
+      "loss": 0.0111,
+      "step": 340
+    },
+    {
+      "epoch": 3.02,
+      "grad_norm": 0.05333692207932472,
+      "learning_rate": 7.370689655172413e-05,
+      "loss": 0.0104,
+      "step": 350
+    },
+    {
+      "epoch": 3.1,
+      "grad_norm": 0.2949971556663513,
+      "learning_rate": 6.724137931034483e-05,
+      "loss": 0.0093,
+      "step": 360
+    },
+    {
+      "epoch": 3.19,
+      "grad_norm": 0.06963898241519928,
+      "learning_rate": 6.077586206896551e-05,
+      "loss": 0.0084,
+      "step": 370
+    },
+    {
+      "epoch": 3.28,
+      "grad_norm": 0.08035852760076523,
+      "learning_rate": 5.4310344827586204e-05,
+      "loss": 0.0067,
+      "step": 380
+    },
+    {
+      "epoch": 3.36,
+      "grad_norm": 0.02499503456056118,
+      "learning_rate": 4.78448275862069e-05,
+      "loss": 0.0034,
+      "step": 390
+    },
+    {
+      "epoch": 3.45,
+      "grad_norm": 0.4326638877391815,
+      "learning_rate": 4.137931034482758e-05,
+      "loss": 0.0087,
+      "step": 400
+    },
+    {
+      "epoch": 3.45,
+      "eval_accuracy": 0.9994587280108255,
+      "eval_loss": 0.0037413176614791155,
+      "eval_runtime": 66.2753,
+      "eval_samples_per_second": 111.505,
+      "eval_steps_per_second": 13.942,
+      "step": 400
+    },
+    {
+      "epoch": 3.53,
+      "grad_norm": 0.016918424516916275,
+      "learning_rate": 3.4913793103448275e-05,
+      "loss": 0.0035,
+      "step": 410
+    },
+    {
+      "epoch": 3.62,
+      "grad_norm": 0.022589026018977165,
+      "learning_rate": 2.8448275862068963e-05,
+      "loss": 0.004,
+      "step": 420
+    },
+    {
+      "epoch": 3.71,
+      "grad_norm": 0.06562381237745285,
+      "learning_rate": 2.198275862068965e-05,
+      "loss": 0.0027,
+      "step": 430
+    },
+    {
+      "epoch": 3.79,
+      "grad_norm": 0.0264147762209177,
+      "learning_rate": 1.5517241379310342e-05,
+      "loss": 0.0027,
+      "step": 440
+    },
+    {
+      "epoch": 3.88,
+      "grad_norm": 0.02782212570309639,
+      "learning_rate": 9.051724137931034e-06,
+      "loss": 0.0046,
+      "step": 450
+    },
+    {
+      "epoch": 3.97,
+      "grad_norm": 0.02095525525510311,
+      "learning_rate": 2.5862068965517237e-06,
+      "loss": 0.0036,
+      "step": 460
+    },
+    {
+      "epoch": 4.0,
+      "step": 464,
+      "total_flos": 2.2913817801515827e+18,
+      "train_loss": 0.18540822506223903,
+      "train_runtime": 740.8045,
+      "train_samples_per_second": 39.903,
+      "train_steps_per_second": 0.626
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 464,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 100,
+  "total_flos": 2.2913817801515827e+18,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72d067b80968c329a6d7da9dee207c0df97fc6614df48511b3b0bb0ba0e75328
+size 4920