🍻 cheers

README.md

@@ -2,6 +2,7 @@
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
+- image-classification
 - generated_from_trainer
 datasets:
 - imagefolder
@@ -14,7 +15,7 @@ model-index:
       name: Image Classification
       type: image-classification
     dataset:
-      name: imagefolder
+      name: action_class
       type: imagefolder
       config: default
       split: train
@@ -30,7 +31,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Action_model
 
-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 imagefolder dataset.
+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 action_class dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.7479
 - Accuracy: 0.8095

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 2.0,
-    "eval_accuracy": 0.7885714285714286,
-    "eval_loss": 0.6974296569824219,
-    "eval_runtime": 15.598,
-    "eval_samples_per_second": 67.316,
-    "eval_steps_per_second": 8.463,
-    "total_flos": 1.555375746295849e+18,
-    "train_loss": 0.33379530375170857,
-    "train_runtime": 590.5675,
-    "train_samples_per_second": 33.984,
-    "train_steps_per_second": 1.063
+    "epoch": 1.0,
+    "eval_accuracy": 0.8095238095238095,
+    "eval_loss": 0.747850775718689,
+    "eval_runtime": 14.6564,
+    "eval_samples_per_second": 71.641,
+    "eval_steps_per_second": 9.006,
+    "total_flos": 7.776878731479245e+17,
+    "train_loss": 1.0910153146002703,
+    "train_runtime": 357.1216,
+    "train_samples_per_second": 28.1,
+    "train_steps_per_second": 0.879
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "eval_accuracy": 0.7885714285714286,
-    "eval_loss": 0.6974296569824219,
-    "eval_runtime": 15.598,
-    "eval_samples_per_second": 67.316,
-    "eval_steps_per_second": 8.463
+    "epoch": 1.0,
+    "eval_accuracy": 0.8095238095238095,
+    "eval_loss": 0.747850775718689,
+    "eval_runtime": 14.6564,
+    "eval_samples_per_second": 71.641,
+    "eval_steps_per_second": 9.006
 }

runs/Apr16_13-47-24_a962527840ae/events.out.tfevents.1713275740.a962527840ae.34.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c2014389d8906c0e5e1087f2252c30400d17433e90d1c62c17dfd64c22e33f3
+size 411

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 2.0,
-    "total_flos": 1.555375746295849e+18,
-    "train_loss": 0.33379530375170857,
-    "train_runtime": 590.5675,
-    "train_samples_per_second": 33.984,
-    "train_steps_per_second": 1.063
+    "epoch": 1.0,
+    "total_flos": 7.776878731479245e+17,
+    "train_loss": 1.0910153146002703,
+    "train_runtime": 357.1216,
+    "train_samples_per_second": 28.1,
+    "train_steps_per_second": 0.879
 }

trainer_state.json

@@ -1,300 +1,161 @@
 {
-  "best_metric": 0.6974296569824219,
-  "best_model_checkpoint": "Action_model/checkpoint-200",
-  "epoch": 2.0,
+  "best_metric": 0.747850775718689,
+  "best_model_checkpoint": "Action_model/checkpoint-300",
+  "epoch": 1.0,
   "eval_steps": 100,
-  "global_step": 628,
+  "global_step": 314,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.06,
-      "grad_norm": 4.470834732055664,
-      "learning_rate": 9.681528662420382e-05,
-      "loss": 0.4514,
+      "grad_norm": 1.3472708463668823,
+      "learning_rate": 9.363057324840766e-05,
+      "loss": 2.1856,
       "step": 20
     },
     {
       "epoch": 0.13,
-      "grad_norm": 3.164530038833618,
-      "learning_rate": 9.363057324840766e-05,
-      "loss": 0.4578,
+      "grad_norm": 1.6540354490280151,
+      "learning_rate": 8.726114649681529e-05,
+      "loss": 1.9121,
       "step": 40
     },
     {
       "epoch": 0.19,
-      "grad_norm": 5.622269630432129,
-      "learning_rate": 9.044585987261147e-05,
-      "loss": 0.3927,
+      "grad_norm": 1.9076273441314697,
+      "learning_rate": 8.089171974522294e-05,
+      "loss": 1.5695,
       "step": 60
     },
     {
       "epoch": 0.25,
-      "grad_norm": 4.455988883972168,
-      "learning_rate": 8.726114649681529e-05,
-      "loss": 0.4274,
+      "grad_norm": 2.171828508377075,
+      "learning_rate": 7.452229299363057e-05,
+      "loss": 1.3273,
       "step": 80
     },
     {
       "epoch": 0.32,
-      "grad_norm": 4.014460563659668,
-      "learning_rate": 8.407643312101911e-05,
-      "loss": 0.4296,
+      "grad_norm": 3.170124053955078,
+      "learning_rate": 6.815286624203822e-05,
+      "loss": 1.1778,
       "step": 100
     },
     {
       "epoch": 0.32,
-      "eval_accuracy": 0.7447619047619047,
-      "eval_loss": 0.8158602118492126,
-      "eval_runtime": 15.7498,
-      "eval_samples_per_second": 66.668,
-      "eval_steps_per_second": 8.381,
+      "eval_accuracy": 0.78,
+      "eval_loss": 1.0651038885116577,
+      "eval_runtime": 20.8448,
+      "eval_samples_per_second": 50.372,
+      "eval_steps_per_second": 6.333,
       "step": 100
     },
     {
       "epoch": 0.38,
-      "grad_norm": 5.374497890472412,
-      "learning_rate": 8.089171974522294e-05,
-      "loss": 0.4462,
+      "grad_norm": 2.705270767211914,
+      "learning_rate": 6.178343949044585e-05,
+      "loss": 1.0757,
       "step": 120
     },
     {
       "epoch": 0.45,
-      "grad_norm": 5.2211833000183105,
-      "learning_rate": 7.786624203821657e-05,
-      "loss": 0.4788,
+      "grad_norm": 2.980999231338501,
+      "learning_rate": 5.5414012738853505e-05,
+      "loss": 1.0115,
       "step": 140
     },
     {
       "epoch": 0.51,
-      "grad_norm": 2.162199020385742,
-      "learning_rate": 7.468152866242038e-05,
-      "loss": 0.3794,
+      "grad_norm": 4.483489990234375,
+      "learning_rate": 4.904458598726115e-05,
+      "loss": 0.9806,
       "step": 160
     },
     {
       "epoch": 0.57,
-      "grad_norm": 2.3695385456085205,
-      "learning_rate": 7.14968152866242e-05,
-      "loss": 0.3611,
+      "grad_norm": 3.3504586219787598,
+      "learning_rate": 4.267515923566879e-05,
+      "loss": 0.8805,
       "step": 180
     },
     {
       "epoch": 0.64,
-      "grad_norm": 3.2987468242645264,
-      "learning_rate": 6.831210191082804e-05,
-      "loss": 0.4083,
+      "grad_norm": 3.248554229736328,
+      "learning_rate": 3.630573248407643e-05,
+      "loss": 0.8527,
       "step": 200
     },
     {
       "epoch": 0.64,
-      "eval_accuracy": 0.7885714285714286,
-      "eval_loss": 0.6974296569824219,
-      "eval_runtime": 14.5668,
-      "eval_samples_per_second": 72.082,
-      "eval_steps_per_second": 9.062,
+      "eval_accuracy": 0.8133333333333334,
+      "eval_loss": 0.7873561978340149,
+      "eval_runtime": 14.6757,
+      "eval_samples_per_second": 71.547,
+      "eval_steps_per_second": 8.994,
       "step": 200
     },
     {
       "epoch": 0.7,
-      "grad_norm": 3.8996469974517822,
-      "learning_rate": 6.512738853503185e-05,
-      "loss": 0.4064,
+      "grad_norm": 2.6971187591552734,
+      "learning_rate": 2.9936305732484078e-05,
+      "loss": 0.7773,
       "step": 220
     },
     {
       "epoch": 0.76,
-      "grad_norm": 4.896476745605469,
-      "learning_rate": 6.194267515923567e-05,
-      "loss": 0.3512,
+      "grad_norm": 2.3084142208099365,
+      "learning_rate": 2.356687898089172e-05,
+      "loss": 0.6965,
       "step": 240
     },
     {
       "epoch": 0.83,
-      "grad_norm": 5.791750907897949,
-      "learning_rate": 5.87579617834395e-05,
-      "loss": 0.411,
+      "grad_norm": 3.3731772899627686,
+      "learning_rate": 1.7197452229299362e-05,
+      "loss": 0.77,
       "step": 260
     },
     {
       "epoch": 0.89,
-      "grad_norm": 4.586860179901123,
-      "learning_rate": 5.5573248407643317e-05,
-      "loss": 0.2974,
+      "grad_norm": 1.8178930282592773,
+      "learning_rate": 1.0828025477707008e-05,
+      "loss": 0.7106,
       "step": 280
     },
     {
       "epoch": 0.96,
-      "grad_norm": 4.982791423797607,
-      "learning_rate": 5.238853503184714e-05,
-      "loss": 0.3384,
+      "grad_norm": 2.819934368133545,
+      "learning_rate": 4.45859872611465e-06,
+      "loss": 0.7114,
       "step": 300
     },
     {
       "epoch": 0.96,
-      "eval_accuracy": 0.7371428571428571,
-      "eval_loss": 0.8742781281471252,
-      "eval_runtime": 14.7081,
-      "eval_samples_per_second": 71.389,
-      "eval_steps_per_second": 8.975,
+      "eval_accuracy": 0.8095238095238095,
+      "eval_loss": 0.747850775718689,
+      "eval_runtime": 14.738,
+      "eval_samples_per_second": 71.244,
+      "eval_steps_per_second": 8.956,
       "step": 300
     },
     {
-      "epoch": 1.02,
-      "grad_norm": 2.271329879760742,
-      "learning_rate": 4.920382165605096e-05,
-      "loss": 0.3209,
-      "step": 320
-    },
-    {
-      "epoch": 1.08,
-      "grad_norm": 4.398645401000977,
-      "learning_rate": 4.601910828025478e-05,
-      "loss": 0.2851,
-      "step": 340
-    },
-    {
-      "epoch": 1.15,
-      "grad_norm": 4.354762077331543,
-      "learning_rate": 4.2834394904458604e-05,
-      "loss": 0.3029,
-      "step": 360
-    },
-    {
-      "epoch": 1.21,
-      "grad_norm": 1.9630823135375977,
-      "learning_rate": 3.964968152866242e-05,
-      "loss": 0.2774,
-      "step": 380
-    },
-    {
-      "epoch": 1.27,
-      "grad_norm": 2.4799087047576904,
-      "learning_rate": 3.646496815286624e-05,
-      "loss": 0.2812,
-      "step": 400
-    },
-    {
-      "epoch": 1.27,
-      "eval_accuracy": 0.7676190476190476,
-      "eval_loss": 0.7738763689994812,
-      "eval_runtime": 14.6761,
-      "eval_samples_per_second": 71.545,
-      "eval_steps_per_second": 8.994,
-      "step": 400
-    },
-    {
-      "epoch": 1.34,
-      "grad_norm": 5.377834320068359,
-      "learning_rate": 3.328025477707007e-05,
-      "loss": 0.2768,
-      "step": 420
-    },
-    {
-      "epoch": 1.4,
-      "grad_norm": 4.753965854644775,
-      "learning_rate": 3.0095541401273885e-05,
-      "loss": 0.3242,
-      "step": 440
-    },
-    {
-      "epoch": 1.46,
-      "grad_norm": 4.039095401763916,
-      "learning_rate": 2.6910828025477707e-05,
-      "loss": 0.2168,
-      "step": 460
-    },
-    {
-      "epoch": 1.53,
-      "grad_norm": 3.443063497543335,
-      "learning_rate": 2.372611464968153e-05,
-      "loss": 0.2584,
-      "step": 480
-    },
-    {
-      "epoch": 1.59,
-      "grad_norm": 1.6640291213989258,
-      "learning_rate": 2.054140127388535e-05,
-      "loss": 0.251,
-      "step": 500
-    },
-    {
-      "epoch": 1.59,
-      "eval_accuracy": 0.7857142857142857,
-      "eval_loss": 0.7165716886520386,
-      "eval_runtime": 14.6301,
-      "eval_samples_per_second": 71.77,
-      "eval_steps_per_second": 9.022,
-      "step": 500
-    },
-    {
-      "epoch": 1.66,
-      "grad_norm": 3.264369487762451,
-      "learning_rate": 1.7356687898089173e-05,
-      "loss": 0.2804,
-      "step": 520
-    },
-    {
-      "epoch": 1.72,
-      "grad_norm": 3.547678232192993,
-      "learning_rate": 1.4171974522292993e-05,
-      "loss": 0.2737,
-      "step": 540
-    },
-    {
-      "epoch": 1.78,
-      "grad_norm": 5.212786674499512,
-      "learning_rate": 1.0987261146496815e-05,
-      "loss": 0.2503,
-      "step": 560
-    },
-    {
-      "epoch": 1.85,
-      "grad_norm": 3.4162983894348145,
-      "learning_rate": 7.802547770700637e-06,
-      "loss": 0.253,
-      "step": 580
-    },
-    {
-      "epoch": 1.91,
-      "grad_norm": 5.042104244232178,
-      "learning_rate": 4.6178343949044585e-06,
-      "loss": 0.2425,
-      "step": 600
-    },
-    {
-      "epoch": 1.91,
-      "eval_accuracy": 0.7742857142857142,
-      "eval_loss": 0.7691510915756226,
-      "eval_runtime": 14.587,
-      "eval_samples_per_second": 71.982,
-      "eval_steps_per_second": 9.049,
-      "step": 600
-    },
-    {
-      "epoch": 1.97,
-      "grad_norm": 5.002872467041016,
-      "learning_rate": 1.4331210191082802e-06,
-      "loss": 0.2493,
-      "step": 620
-    },
-    {
-      "epoch": 2.0,
-      "step": 628,
-      "total_flos": 1.555375746295849e+18,
-      "train_loss": 0.33379530375170857,
-      "train_runtime": 590.5675,
-      "train_samples_per_second": 33.984,
-      "train_steps_per_second": 1.063
+      "epoch": 1.0,
+      "step": 314,
+      "total_flos": 7.776878731479245e+17,
+      "train_loss": 1.0910153146002703,
+      "train_runtime": 357.1216,
+      "train_samples_per_second": 28.1,
+      "train_steps_per_second": 0.879
     }
   ],
   "logging_steps": 20,
-  "max_steps": 628,
+  "max_steps": 314,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 2,
+  "num_train_epochs": 1,
   "save_steps": 100,
-  "total_flos": 1.555375746295849e+18,
+  "total_flos": 7.776878731479245e+17,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null