End of training

README.md

@@ -2,6 +2,8 @@
 license: apache-2.0
 base_model: google/vit-large-patch16-224-in21k
 tags:
+- image-classification
+- vision
 - generated_from_trainer
 datasets:
 - imagefolder
@@ -14,7 +16,7 @@ model-index:
       name: Image Classification
       type: image-classification
     dataset:
-      name: imagefolder
+      name: touchtech/fashion-images-perspectives
       type: imagefolder
       config: default
       split: train
@@ -30,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # fashion-images-perspectives-vit-large-patch16-224-in21k-v4
 
-This model is a fine-tuned version of [google/vit-large-patch16-224-in21k](https://huggingface.co/google/vit-large-patch16-224-in21k) on the imagefolder dataset.
+This model is a fine-tuned version of [google/vit-large-patch16-224-in21k](https://huggingface.co/google/vit-large-patch16-224-in21k) on the touchtech/fashion-images-perspectives dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.2203
 - Accuracy: 0.9434

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.9434482758620689,
+    "eval_loss": 0.22029058635234833,
+    "eval_runtime": 314.6406,
+    "eval_samples_per_second": 13.825,
+    "eval_steps_per_second": 1.729,
+    "train_loss": 0.33228889720388066,
+    "train_runtime": 12005.4433,
+    "train_samples_per_second": 10.264,
+    "train_steps_per_second": 1.283
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.9434482758620689,
+    "eval_loss": 0.22029058635234833,
+    "eval_runtime": 314.6406,
+    "eval_samples_per_second": 13.825,
+    "eval_steps_per_second": 1.729
+}

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 5.0,
+    "train_loss": 0.33228889720388066,
+    "train_runtime": 12005.4433,
+    "train_samples_per_second": 10.264,
+    "train_steps_per_second": 1.283
+}

trainer_state.json

@@ -0,0 +1,253 @@
+{
+  "best_metric": 0.22029058635234833,
+  "best_model_checkpoint": "/workspace/training_output/perspectives-vit-large-patch16-224-in21k-v4/checkpoint-15405",
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 15405,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.9350860110353782e-05,
+      "loss": 0.887,
+      "step": 500
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.8701720220707566e-05,
+      "loss": 0.5859,
+      "step": 1000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 1.8052580331061346e-05,
+      "loss": 0.5192,
+      "step": 1500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 1.7403440441415126e-05,
+      "loss": 0.4758,
+      "step": 2000
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.6754300551768907e-05,
+      "loss": 0.4477,
+      "step": 2500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 1.610516066212269e-05,
+      "loss": 0.4275,
+      "step": 3000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9011494252873563,
+      "eval_loss": 0.306373655796051,
+      "eval_runtime": 350.0887,
+      "eval_samples_per_second": 12.425,
+      "eval_steps_per_second": 1.554,
+      "step": 3081
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.545602077247647e-05,
+      "loss": 0.3785,
+      "step": 3500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.480688088283025e-05,
+      "loss": 0.3531,
+      "step": 4000
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 1.4157740993184033e-05,
+      "loss": 0.3636,
+      "step": 4500
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 1.3508601103537813e-05,
+      "loss": 0.3663,
+      "step": 5000
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.2859461213891595e-05,
+      "loss": 0.345,
+      "step": 5500
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 1.2210321324245376e-05,
+      "loss": 0.3555,
+      "step": 6000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9103448275862069,
+      "eval_loss": 0.3097192347049713,
+      "eval_runtime": 317.5331,
+      "eval_samples_per_second": 13.699,
+      "eval_steps_per_second": 1.713,
+      "step": 6162
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 1.1561181434599158e-05,
+      "loss": 0.2992,
+      "step": 6500
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 1.0912041544952938e-05,
+      "loss": 0.3178,
+      "step": 7000
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 1.0262901655306718e-05,
+      "loss": 0.3011,
+      "step": 7500
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 9.6137617656605e-06,
+      "loss": 0.2823,
+      "step": 8000
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 8.964621876014283e-06,
+      "loss": 0.2864,
+      "step": 8500
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 8.315481986368063e-06,
+      "loss": 0.3069,
+      "step": 9000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9105747126436782,
+      "eval_loss": 0.30360889434814453,
+      "eval_runtime": 336.8305,
+      "eval_samples_per_second": 12.915,
+      "eval_steps_per_second": 1.615,
+      "step": 9243
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 7.666342096721843e-06,
+      "loss": 0.2615,
+      "step": 9500
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 7.017202207075625e-06,
+      "loss": 0.2504,
+      "step": 10000
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 6.368062317429406e-06,
+      "loss": 0.247,
+      "step": 10500
+    },
+    {
+      "epoch": 3.57,
+      "learning_rate": 5.7189224277831876e-06,
+      "loss": 0.2531,
+      "step": 11000
+    },
+    {
+      "epoch": 3.73,
+      "learning_rate": 5.06978253813697e-06,
+      "loss": 0.222,
+      "step": 11500
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 4.42064264849075e-06,
+      "loss": 0.2449,
+      "step": 12000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9377011494252874,
+      "eval_loss": 0.22682134807109833,
+      "eval_runtime": 316.4362,
+      "eval_samples_per_second": 13.747,
+      "eval_steps_per_second": 1.719,
+      "step": 12324
+    },
+    {
+      "epoch": 4.06,
+      "learning_rate": 3.7715027588445315e-06,
+      "loss": 0.2261,
+      "step": 12500
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 3.1223628691983127e-06,
+      "loss": 0.221,
+      "step": 13000
+    },
+    {
+      "epoch": 4.38,
+      "learning_rate": 2.473222979552094e-06,
+      "loss": 0.2247,
+      "step": 13500
+    },
+    {
+      "epoch": 4.54,
+      "learning_rate": 1.8240830899058748e-06,
+      "loss": 0.2037,
+      "step": 14000
+    },
+    {
+      "epoch": 4.71,
+      "learning_rate": 1.174943200259656e-06,
+      "loss": 0.2023,
+      "step": 14500
+    },
+    {
+      "epoch": 4.87,
+      "learning_rate": 5.258033106134373e-07,
+      "loss": 0.2339,
+      "step": 15000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9434482758620689,
+      "eval_loss": 0.22029058635234833,
+      "eval_runtime": 317.0472,
+      "eval_samples_per_second": 13.72,
+      "eval_steps_per_second": 1.716,
+      "step": 15405
+    },
+    {
+      "epoch": 5.0,
+      "step": 15405,
+      "total_flos": 3.3756908216197018e+19,
+      "train_loss": 0.33228889720388066,
+      "train_runtime": 12005.4433,
+      "train_samples_per_second": 10.264,
+      "train_steps_per_second": 1.283
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 15405,
+  "num_train_epochs": 5,
+  "save_steps": 500,
+  "total_flos": 3.3756908216197018e+19,
+  "trial_name": null,
+  "trial_params": null
+}