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-perspectives-v2
       type: imagefolder
       config: default
       split: train
@@ -23,7 +25,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.9641970198675497
+      value: 0.9544701986754967
 ---
 
 <!-- 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-perspectives-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-perspectives-v2 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2290
-- Accuracy: 0.9642
+- Loss: 0.1572
+- Accuracy: 0.9545
 
 ## Model description
 

all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.9544701986754967,
+    "eval_loss": 0.15720506012439728,
+    "eval_runtime": 447.4514,
+    "eval_samples_per_second": 10.799,
+    "eval_steps_per_second": 1.35,
+    "total_flos": 1.1034252095877513e+20,
+    "train_loss": 0.08469340640865429,
+    "train_runtime": 22735.4231,
+    "train_samples_per_second": 6.021,
+    "train_steps_per_second": 0.753
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "eval_accuracy": 0.9544701986754967,
+    "eval_loss": 0.15720506012439728,
+    "eval_runtime": 447.4514,
+    "eval_samples_per_second": 10.799,
+    "eval_steps_per_second": 1.35
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "total_flos": 1.1034252095877513e+20,
+    "train_loss": 0.08469340640865429,
+    "train_runtime": 22735.4231,
+    "train_samples_per_second": 6.021,
+    "train_steps_per_second": 0.753
+}

trainer_state.json

@@ -0,0 +1,325 @@
+{
+  "best_metric": 0.15720506012439728,
+  "best_model_checkpoint": "/workspace/training_output/perspectives-vit-large-patch16-384-v1/checkpoint-3423",
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 17115,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.14607069821793747,
+      "grad_norm": 2.9760661125183105,
+      "learning_rate": 1.941571720712825e-05,
+      "loss": 0.4878,
+      "step": 500
+    },
+    {
+      "epoch": 0.29214139643587494,
+      "grad_norm": 0.5538421869277954,
+      "learning_rate": 1.8831434414256503e-05,
+      "loss": 0.2583,
+      "step": 1000
+    },
+    {
+      "epoch": 0.43821209465381245,
+      "grad_norm": 8.382941246032715,
+      "learning_rate": 1.8247151621384752e-05,
+      "loss": 0.2558,
+      "step": 1500
+    },
+    {
+      "epoch": 0.5842827928717499,
+      "grad_norm": 2.5761232376098633,
+      "learning_rate": 1.7662868828513e-05,
+      "loss": 0.2137,
+      "step": 2000
+    },
+    {
+      "epoch": 0.7303534910896874,
+      "grad_norm": 8.756667137145996,
+      "learning_rate": 1.707858603564125e-05,
+      "loss": 0.1906,
+      "step": 2500
+    },
+    {
+      "epoch": 0.8764241893076249,
+      "grad_norm": 0.7560040354728699,
+      "learning_rate": 1.6494303242769503e-05,
+      "loss": 0.1766,
+      "step": 3000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9544701986754967,
+      "eval_loss": 0.15720506012439728,
+      "eval_runtime": 454.0694,
+      "eval_samples_per_second": 10.642,
+      "eval_steps_per_second": 1.33,
+      "step": 3423
+    },
+    {
+      "epoch": 1.0224948875255624,
+      "grad_norm": 0.2756825089454651,
+      "learning_rate": 1.591002044989775e-05,
+      "loss": 0.1593,
+      "step": 3500
+    },
+    {
+      "epoch": 1.1685655857434998,
+      "grad_norm": 14.493992805480957,
+      "learning_rate": 1.5325737657026e-05,
+      "loss": 0.0946,
+      "step": 4000
+    },
+    {
+      "epoch": 1.3146362839614374,
+      "grad_norm": 1.4798998832702637,
+      "learning_rate": 1.4741454864154251e-05,
+      "loss": 0.096,
+      "step": 4500
+    },
+    {
+      "epoch": 1.4607069821793748,
+      "grad_norm": 0.03895518556237221,
+      "learning_rate": 1.41571720712825e-05,
+      "loss": 0.1097,
+      "step": 5000
+    },
+    {
+      "epoch": 1.6067776803973124,
+      "grad_norm": 0.019592493772506714,
+      "learning_rate": 1.3572889278410753e-05,
+      "loss": 0.1052,
+      "step": 5500
+    },
+    {
+      "epoch": 1.7528483786152498,
+      "grad_norm": 0.019370147958397865,
+      "learning_rate": 1.2988606485539002e-05,
+      "loss": 0.115,
+      "step": 6000
+    },
+    {
+      "epoch": 1.8989190768331872,
+      "grad_norm": 2.880488634109497,
+      "learning_rate": 1.2404323692667253e-05,
+      "loss": 0.1108,
+      "step": 6500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9579884105960265,
+      "eval_loss": 0.1967967003583908,
+      "eval_runtime": 457.9224,
+      "eval_samples_per_second": 10.552,
+      "eval_steps_per_second": 1.319,
+      "step": 6846
+    },
+    {
+      "epoch": 2.044989775051125,
+      "grad_norm": 0.12547972798347473,
+      "learning_rate": 1.1820040899795502e-05,
+      "loss": 0.0913,
+      "step": 7000
+    },
+    {
+      "epoch": 2.1910604732690624,
+      "grad_norm": 0.01506296917796135,
+      "learning_rate": 1.123575810692375e-05,
+      "loss": 0.0423,
+      "step": 7500
+    },
+    {
+      "epoch": 2.3371311714869996,
+      "grad_norm": 19.676223754882812,
+      "learning_rate": 1.0651475314052001e-05,
+      "loss": 0.048,
+      "step": 8000
+    },
+    {
+      "epoch": 2.483201869704937,
+      "grad_norm": 0.19931885600090027,
+      "learning_rate": 1.006719252118025e-05,
+      "loss": 0.0416,
+      "step": 8500
+    },
+    {
+      "epoch": 2.629272567922875,
+      "grad_norm": 0.01677100360393524,
+      "learning_rate": 9.482909728308503e-06,
+      "loss": 0.0493,
+      "step": 9000
+    },
+    {
+      "epoch": 2.775343266140812,
+      "grad_norm": 0.008517206646502018,
+      "learning_rate": 8.898626935436752e-06,
+      "loss": 0.044,
+      "step": 9500
+    },
+    {
+      "epoch": 2.9214139643587496,
+      "grad_norm": 0.006620448548346758,
+      "learning_rate": 8.314344142565001e-06,
+      "loss": 0.0456,
+      "step": 10000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9606788079470199,
+      "eval_loss": 0.2058100551366806,
+      "eval_runtime": 454.6414,
+      "eval_samples_per_second": 10.628,
+      "eval_steps_per_second": 1.329,
+      "step": 10269
+    },
+    {
+      "epoch": 3.067484662576687,
+      "grad_norm": 0.0024903384037315845,
+      "learning_rate": 7.730061349693252e-06,
+      "loss": 0.0234,
+      "step": 10500
+    },
+    {
+      "epoch": 3.213555360794625,
+      "grad_norm": 0.015993278473615646,
+      "learning_rate": 7.1457785568215025e-06,
+      "loss": 0.0075,
+      "step": 11000
+    },
+    {
+      "epoch": 3.359626059012562,
+      "grad_norm": 0.027190232649445534,
+      "learning_rate": 6.561495763949752e-06,
+      "loss": 0.0112,
+      "step": 11500
+    },
+    {
+      "epoch": 3.5056967572304996,
+      "grad_norm": 0.010302845388650894,
+      "learning_rate": 5.977212971078002e-06,
+      "loss": 0.0188,
+      "step": 12000
+    },
+    {
+      "epoch": 3.651767455448437,
+      "grad_norm": 0.0008801922085694969,
+      "learning_rate": 5.392930178206253e-06,
+      "loss": 0.0189,
+      "step": 12500
+    },
+    {
+      "epoch": 3.7978381536663743,
+      "grad_norm": 12.479740142822266,
+      "learning_rate": 4.808647385334502e-06,
+      "loss": 0.025,
+      "step": 13000
+    },
+    {
+      "epoch": 3.943908851884312,
+      "grad_norm": 0.0013448131503537297,
+      "learning_rate": 4.224364592462753e-06,
+      "loss": 0.0118,
+      "step": 13500
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.9633692052980133,
+      "eval_loss": 0.2165926992893219,
+      "eval_runtime": 447.8313,
+      "eval_samples_per_second": 10.79,
+      "eval_steps_per_second": 1.349,
+      "step": 13692
+    },
+    {
+      "epoch": 4.08997955010225,
+      "grad_norm": 0.0009766396833583713,
+      "learning_rate": 3.6400817995910027e-06,
+      "loss": 0.012,
+      "step": 14000
+    },
+    {
+      "epoch": 4.236050248320187,
+      "grad_norm": 0.0005415708874352276,
+      "learning_rate": 3.055799006719252e-06,
+      "loss": 0.0037,
+      "step": 14500
+    },
+    {
+      "epoch": 4.382120946538125,
+      "grad_norm": 0.0017334007425233722,
+      "learning_rate": 2.4715162138475024e-06,
+      "loss": 0.007,
+      "step": 15000
+    },
+    {
+      "epoch": 4.5281916447560615,
+      "grad_norm": 0.00246521458029747,
+      "learning_rate": 1.8872334209757523e-06,
+      "loss": 0.0058,
+      "step": 15500
+    },
+    {
+      "epoch": 4.674262342973999,
+      "grad_norm": 0.0018645540112629533,
+      "learning_rate": 1.3029506281040024e-06,
+      "loss": 0.0066,
+      "step": 16000
+    },
+    {
+      "epoch": 4.820333041191937,
+      "grad_norm": 0.0014726301888003945,
+      "learning_rate": 7.186678352322525e-07,
+      "loss": 0.0047,
+      "step": 16500
+    },
+    {
+      "epoch": 4.966403739409874,
+      "grad_norm": 0.0010302780428901315,
+      "learning_rate": 1.3438504236050248e-07,
+      "loss": 0.007,
+      "step": 17000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.9641970198675497,
+      "eval_loss": 0.22900521755218506,
+      "eval_runtime": 454.8427,
+      "eval_samples_per_second": 10.623,
+      "eval_steps_per_second": 1.328,
+      "step": 17115
+    },
+    {
+      "epoch": 5.0,
+      "step": 17115,
+      "total_flos": 1.1034252095877513e+20,
+      "train_loss": 0.08469340640865429,
+      "train_runtime": 22735.4231,
+      "train_samples_per_second": 6.021,
+      "train_steps_per_second": 0.753
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 17115,
+  "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": 1.1034252095877513e+20,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}