End of training

README.md

@@ -3,6 +3,8 @@ library_name: transformers
 license: apache-2.0
 base_model: google/vit-large-patch16-384
 tags:
+- image-classification
+- vision
 - generated_from_trainer
 datasets:
 - imagefolder
@@ -15,7 +17,7 @@ model-index:
       name: Image Classification
       type: image-classification
     dataset:
-      name: imagefolder
+      name: touchtech/fashion-images-pack-types
       type: imagefolder
       config: default
       split: train
@@ -23,7 +25,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.991652754590985
+      value: 0.989983305509182
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -31,10 +33,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # fashion-images-pack-types-vit-large-patch16-384-v1
 
-This model is a fine-tuned version of [google/vit-large-patch16-384](https://huggingface.co/google/vit-large-patch16-384) on the imagefolder dataset.
+This model is a fine-tuned version of [google/vit-large-patch16-384](https://huggingface.co/google/vit-large-patch16-384) on the touchtech/fashion-images-pack-types dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0485
-- Accuracy: 0.9917
+- Loss: 0.0446
+- Accuracy: 0.9900
 
 ## Model description
 

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.989983305509182,
+    "eval_loss": 0.044622376561164856,
+    "eval_runtime": 212.2261,
+    "eval_samples_per_second": 11.29,
+    "eval_steps_per_second": 1.414,
+    "total_flos": 5.471324673450394e+19,
+    "train_loss": 0.02982461175216723,
+    "train_runtime": 10814.9254,
+    "train_samples_per_second": 6.276,
+    "train_steps_per_second": 0.785
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.989983305509182,
+    "eval_loss": 0.044622376561164856,
+    "eval_runtime": 212.2261,
+    "eval_samples_per_second": 11.29,
+    "eval_steps_per_second": 1.414
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "total_flos": 5.471324673450394e+19,
+    "train_loss": 0.02982461175216723,
+    "train_runtime": 10814.9254,
+    "train_samples_per_second": 6.276,
+    "train_steps_per_second": 0.785
+}

trainer_state.json

@@ -0,0 +1,199 @@
+{
+  "best_metric": 0.044622376561164856,
+  "best_model_checkpoint": "/training_output/fashion-images-pack-types-vit-large-patch16-384-v1/checkpoint-5091",
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 8485,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.2946375957572186,
+      "grad_norm": 0.6524115204811096,
+      "learning_rate": 1.8821449616971128e-05,
+      "loss": 0.1857,
+      "step": 500
+    },
+    {
+      "epoch": 0.5892751915144372,
+      "grad_norm": 0.0360204242169857,
+      "learning_rate": 1.764289923394225e-05,
+      "loss": 0.0812,
+      "step": 1000
+    },
+    {
+      "epoch": 0.8839127872716559,
+      "grad_norm": 0.011747363954782486,
+      "learning_rate": 1.646434885091338e-05,
+      "loss": 0.0823,
+      "step": 1500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9837228714524207,
+      "eval_loss": 0.0661115050315857,
+      "eval_runtime": 214.9346,
+      "eval_samples_per_second": 11.148,
+      "eval_steps_per_second": 1.396,
+      "step": 1697
+    },
+    {
+      "epoch": 1.1785503830288744,
+      "grad_norm": 2.4793760776519775,
+      "learning_rate": 1.5285798467884503e-05,
+      "loss": 0.0427,
+      "step": 2000
+    },
+    {
+      "epoch": 1.473187978786093,
+      "grad_norm": 0.018900904804468155,
+      "learning_rate": 1.410724808485563e-05,
+      "loss": 0.0208,
+      "step": 2500
+    },
+    {
+      "epoch": 1.7678255745433118,
+      "grad_norm": 0.002555049257352948,
+      "learning_rate": 1.2928697701826754e-05,
+      "loss": 0.0197,
+      "step": 3000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9812186978297162,
+      "eval_loss": 0.08999926596879959,
+      "eval_runtime": 212.7124,
+      "eval_samples_per_second": 11.264,
+      "eval_steps_per_second": 1.41,
+      "step": 3394
+    },
+    {
+      "epoch": 2.0624631703005303,
+      "grad_norm": 0.0002872613840736449,
+      "learning_rate": 1.175014731879788e-05,
+      "loss": 0.0238,
+      "step": 3500
+    },
+    {
+      "epoch": 2.3571007660577488,
+      "grad_norm": 0.060405392199754715,
+      "learning_rate": 1.0571596935769004e-05,
+      "loss": 0.0075,
+      "step": 4000
+    },
+    {
+      "epoch": 2.6517383618149677,
+      "grad_norm": 0.001587436068803072,
+      "learning_rate": 9.39304655274013e-06,
+      "loss": 0.011,
+      "step": 4500
+    },
+    {
+      "epoch": 2.946375957572186,
+      "grad_norm": 0.013891604728996754,
+      "learning_rate": 8.214496169711255e-06,
+      "loss": 0.015,
+      "step": 5000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.989983305509182,
+      "eval_loss": 0.044622376561164856,
+      "eval_runtime": 211.8864,
+      "eval_samples_per_second": 11.308,
+      "eval_steps_per_second": 1.416,
+      "step": 5091
+    },
+    {
+      "epoch": 3.2410135533294047,
+      "grad_norm": 0.0014336752938106656,
+      "learning_rate": 7.035945786682381e-06,
+      "loss": 0.0036,
+      "step": 5500
+    },
+    {
+      "epoch": 3.5356511490866236,
+      "grad_norm": 6.203103112056851e-05,
+      "learning_rate": 5.857395403653507e-06,
+      "loss": 0.0069,
+      "step": 6000
+    },
+    {
+      "epoch": 3.830288744843842,
+      "grad_norm": 0.0005701961345039308,
+      "learning_rate": 4.678845020624632e-06,
+      "loss": 0.0017,
+      "step": 6500
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9912353923205343,
+      "eval_loss": 0.051807090640068054,
+      "eval_runtime": 210.707,
+      "eval_samples_per_second": 11.371,
+      "eval_steps_per_second": 1.424,
+      "step": 6788
+    },
+    {
+      "epoch": 4.124926340601061,
+      "grad_norm": 0.00022500261547975242,
+      "learning_rate": 3.5002946375957573e-06,
+      "loss": 0.002,
+      "step": 7000
+    },
+    {
+      "epoch": 4.419563936358279,
+      "grad_norm": 5.4514002840733156e-05,
+      "learning_rate": 2.3217442545668827e-06,
+      "loss": 0.0021,
+      "step": 7500
+    },
+    {
+      "epoch": 4.7142015321154975,
+      "grad_norm": 0.00018313823966309428,
+      "learning_rate": 1.1431938715380085e-06,
+      "loss": 0.0,
+      "step": 8000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.991652754590985,
+      "eval_loss": 0.048458877950906754,
+      "eval_runtime": 210.4591,
+      "eval_samples_per_second": 11.385,
+      "eval_steps_per_second": 1.425,
+      "step": 8485
+    },
+    {
+      "epoch": 5.0,
+      "step": 8485,
+      "total_flos": 5.471324673450394e+19,
+      "train_loss": 0.02982461175216723,
+      "train_runtime": 10814.9254,
+      "train_samples_per_second": 6.276,
+      "train_steps_per_second": 0.785
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 8485,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "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": 5.471324673450394e+19,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}