🍻 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
@@ -22,7 +23,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.7742857142857142
+      value: 0.7885714285714286
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -30,10 +31,10 @@ 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: 0.7692
-- Accuracy: 0.7743
+- Loss: 0.6974
+- Accuracy: 0.7886
 
 ## Model description
 

all_results.json

@@ -1,8 +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": 0.7493580146959633,
-    "train_runtime": 668.3994,
-    "train_samples_per_second": 30.027,
-    "train_steps_per_second": 0.94
+    "train_loss": 0.33379530375170857,
+    "train_runtime": 590.5675,
+    "train_samples_per_second": 33.984,
+    "train_steps_per_second": 1.063
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "eval_accuracy": 0.799047619047619,
-    "eval_loss": 0.6551458239555359,
-    "eval_runtime": 15.0447,
-    "eval_samples_per_second": 69.792,
-    "eval_steps_per_second": 8.774
+    "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
 }

runs/Apr16_07-14-49_7b4e89625b83/events.out.tfevents.1713255623.7b4e89625b83.34.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1130ab17d2afd6f827305aafdeec78829b8251cf2b664b9ba7b1eb859f6f7aa
+size 411

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.0,
     "total_flos": 1.555375746295849e+18,
-    "train_loss": 0.7493580146959633,
-    "train_runtime": 668.3994,
-    "train_samples_per_second": 30.027,
-    "train_steps_per_second": 0.94
+    "train_loss": 0.33379530375170857,
+    "train_runtime": 590.5675,
+    "train_samples_per_second": 33.984,
+    "train_steps_per_second": 1.063
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.6526193618774414,
-  "best_model_checkpoint": "Action_model/checkpoint-600",
+  "best_metric": 0.6974296569824219,
+  "best_model_checkpoint": "Action_model/checkpoint-200",
   "epoch": 2.0,
   "eval_steps": 100,
   "global_step": 628,
@@ -10,283 +10,283 @@
   "log_history": [
     {
       "epoch": 0.06,
-      "grad_norm": 1.701284408569336,
+      "grad_norm": 4.470834732055664,
       "learning_rate": 9.681528662420382e-05,
-      "loss": 2.1749,
+      "loss": 0.4514,
       "step": 20
     },
     {
       "epoch": 0.13,
-      "grad_norm": 1.666106939315796,
+      "grad_norm": 3.164530038833618,
       "learning_rate": 9.363057324840766e-05,
-      "loss": 1.8733,
+      "loss": 0.4578,
       "step": 40
     },
     {
       "epoch": 0.19,
-      "grad_norm": 2.066211700439453,
+      "grad_norm": 5.622269630432129,
       "learning_rate": 9.044585987261147e-05,
-      "loss": 1.5213,
+      "loss": 0.3927,
       "step": 60
     },
     {
       "epoch": 0.25,
-      "grad_norm": 2.146984815597534,
+      "grad_norm": 4.455988883972168,
       "learning_rate": 8.726114649681529e-05,
-      "loss": 1.2918,
+      "loss": 0.4274,
       "step": 80
     },
     {
       "epoch": 0.32,
-      "grad_norm": 3.542285442352295,
+      "grad_norm": 4.014460563659668,
       "learning_rate": 8.407643312101911e-05,
-      "loss": 1.1323,
+      "loss": 0.4296,
       "step": 100
     },
     {
       "epoch": 0.32,
-      "eval_accuracy": 0.7542857142857143,
-      "eval_loss": 1.0433851480484009,
-      "eval_runtime": 23.4991,
-      "eval_samples_per_second": 44.683,
-      "eval_steps_per_second": 5.617,
+      "eval_accuracy": 0.7447619047619047,
+      "eval_loss": 0.8158602118492126,
+      "eval_runtime": 15.7498,
+      "eval_samples_per_second": 66.668,
+      "eval_steps_per_second": 8.381,
       "step": 100
     },
     {
       "epoch": 0.38,
-      "grad_norm": 3.3140709400177,
+      "grad_norm": 5.374497890472412,
       "learning_rate": 8.089171974522294e-05,
-      "loss": 1.0549,
+      "loss": 0.4462,
       "step": 120
     },
     {
       "epoch": 0.45,
-      "grad_norm": 2.951521158218384,
-      "learning_rate": 7.770700636942676e-05,
-      "loss": 0.9484,
+      "grad_norm": 5.2211833000183105,
+      "learning_rate": 7.786624203821657e-05,
+      "loss": 0.4788,
       "step": 140
     },
     {
       "epoch": 0.51,
-      "grad_norm": 2.615647554397583,
-      "learning_rate": 7.452229299363057e-05,
-      "loss": 0.9485,
+      "grad_norm": 2.162199020385742,
+      "learning_rate": 7.468152866242038e-05,
+      "loss": 0.3794,
       "step": 160
     },
     {
       "epoch": 0.57,
-      "grad_norm": 3.7320423126220703,
-      "learning_rate": 7.13375796178344e-05,
-      "loss": 0.8395,
+      "grad_norm": 2.3695385456085205,
+      "learning_rate": 7.14968152866242e-05,
+      "loss": 0.3611,
       "step": 180
     },
     {
       "epoch": 0.64,
-      "grad_norm": 2.483612060546875,
-      "learning_rate": 6.815286624203822e-05,
-      "loss": 0.7842,
+      "grad_norm": 3.2987468242645264,
+      "learning_rate": 6.831210191082804e-05,
+      "loss": 0.4083,
       "step": 200
     },
     {
       "epoch": 0.64,
-      "eval_accuracy": 0.780952380952381,
-      "eval_loss": 0.7771612405776978,
-      "eval_runtime": 14.6905,
-      "eval_samples_per_second": 71.475,
-      "eval_steps_per_second": 8.985,
+      "eval_accuracy": 0.7885714285714286,
+      "eval_loss": 0.6974296569824219,
+      "eval_runtime": 14.5668,
+      "eval_samples_per_second": 72.082,
+      "eval_steps_per_second": 9.062,
       "step": 200
     },
     {
       "epoch": 0.7,
-      "grad_norm": 2.181293487548828,
+      "grad_norm": 3.8996469974517822,
       "learning_rate": 6.512738853503185e-05,
-      "loss": 0.7154,
+      "loss": 0.4064,
       "step": 220
     },
     {
       "epoch": 0.76,
-      "grad_norm": 3.2365713119506836,
+      "grad_norm": 4.896476745605469,
       "learning_rate": 6.194267515923567e-05,
-      "loss": 0.6236,
+      "loss": 0.3512,
       "step": 240
     },
     {
       "epoch": 0.83,
-      "grad_norm": 3.164695978164673,
+      "grad_norm": 5.791750907897949,
       "learning_rate": 5.87579617834395e-05,
-      "loss": 0.706,
+      "loss": 0.411,
       "step": 260
     },
     {
       "epoch": 0.89,
-      "grad_norm": 3.6914055347442627,
+      "grad_norm": 4.586860179901123,
       "learning_rate": 5.5573248407643317e-05,
-      "loss": 0.6549,
+      "loss": 0.2974,
       "step": 280
     },
     {
       "epoch": 0.96,
-      "grad_norm": 3.365011215209961,
+      "grad_norm": 4.982791423797607,
       "learning_rate": 5.238853503184714e-05,
-      "loss": 0.6174,
+      "loss": 0.3384,
       "step": 300
     },
     {
       "epoch": 0.96,
-      "eval_accuracy": 0.7847619047619048,
-      "eval_loss": 0.7121406197547913,
-      "eval_runtime": 14.6509,
-      "eval_samples_per_second": 71.668,
-      "eval_steps_per_second": 9.01,
+      "eval_accuracy": 0.7371428571428571,
+      "eval_loss": 0.8742781281471252,
+      "eval_runtime": 14.7081,
+      "eval_samples_per_second": 71.389,
+      "eval_steps_per_second": 8.975,
       "step": 300
     },
     {
       "epoch": 1.02,
-      "grad_norm": 3.2638425827026367,
-      "learning_rate": 4.9363057324840765e-05,
-      "loss": 0.5646,
+      "grad_norm": 2.271329879760742,
+      "learning_rate": 4.920382165605096e-05,
+      "loss": 0.3209,
       "step": 320
     },
     {
       "epoch": 1.08,
-      "grad_norm": 4.049094200134277,
-      "learning_rate": 4.617834394904459e-05,
-      "loss": 0.5271,
+      "grad_norm": 4.398645401000977,
+      "learning_rate": 4.601910828025478e-05,
+      "loss": 0.2851,
       "step": 340
     },
     {
       "epoch": 1.15,
-      "grad_norm": 2.9733641147613525,
-      "learning_rate": 4.299363057324841e-05,
-      "loss": 0.5735,
+      "grad_norm": 4.354762077331543,
+      "learning_rate": 4.2834394904458604e-05,
+      "loss": 0.3029,
       "step": 360
     },
     {
       "epoch": 1.21,
-      "grad_norm": 3.511613130569458,
-      "learning_rate": 3.9808917197452234e-05,
-      "loss": 0.5137,
+      "grad_norm": 1.9630823135375977,
+      "learning_rate": 3.964968152866242e-05,
+      "loss": 0.2774,
       "step": 380
     },
     {
       "epoch": 1.27,
-      "grad_norm": 2.1663146018981934,
-      "learning_rate": 3.662420382165605e-05,
-      "loss": 0.5362,
+      "grad_norm": 2.4799087047576904,
+      "learning_rate": 3.646496815286624e-05,
+      "loss": 0.2812,
       "step": 400
     },
     {
       "epoch": 1.27,
-      "eval_accuracy": 0.7723809523809524,
-      "eval_loss": 0.7485681772232056,
-      "eval_runtime": 14.6889,
-      "eval_samples_per_second": 71.483,
-      "eval_steps_per_second": 8.986,
+      "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": 3.5004124641418457,
-      "learning_rate": 3.343949044585987e-05,
-      "loss": 0.468,
+      "grad_norm": 5.377834320068359,
+      "learning_rate": 3.328025477707007e-05,
+      "loss": 0.2768,
       "step": 420
     },
     {
       "epoch": 1.4,
-      "grad_norm": 2.454324722290039,
-      "learning_rate": 3.0254777070063693e-05,
-      "loss": 0.474,
+      "grad_norm": 4.753965854644775,
+      "learning_rate": 3.0095541401273885e-05,
+      "loss": 0.3242,
       "step": 440
     },
     {
       "epoch": 1.46,
-      "grad_norm": 3.5527968406677246,
-      "learning_rate": 2.707006369426752e-05,
-      "loss": 0.3886,
+      "grad_norm": 4.039095401763916,
+      "learning_rate": 2.6910828025477707e-05,
+      "loss": 0.2168,
       "step": 460
     },
     {
       "epoch": 1.53,
-      "grad_norm": 4.704432487487793,
-      "learning_rate": 2.388535031847134e-05,
-      "loss": 0.3968,
+      "grad_norm": 3.443063497543335,
+      "learning_rate": 2.372611464968153e-05,
+      "loss": 0.2584,
       "step": 480
     },
     {
       "epoch": 1.59,
-      "grad_norm": 3.4028120040893555,
-      "learning_rate": 2.0700636942675162e-05,
-      "loss": 0.4918,
+      "grad_norm": 1.6640291213989258,
+      "learning_rate": 2.054140127388535e-05,
+      "loss": 0.251,
       "step": 500
     },
     {
       "epoch": 1.59,
-      "eval_accuracy": 0.8028571428571428,
-      "eval_loss": 0.6674752235412598,
-      "eval_runtime": 14.482,
-      "eval_samples_per_second": 72.504,
-      "eval_steps_per_second": 9.115,
+      "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": 5.370004177093506,
-      "learning_rate": 1.751592356687898e-05,
-      "loss": 0.4256,
+      "grad_norm": 3.264369487762451,
+      "learning_rate": 1.7356687898089173e-05,
+      "loss": 0.2804,
       "step": 520
     },
     {
       "epoch": 1.72,
-      "grad_norm": 3.2904512882232666,
-      "learning_rate": 1.4490445859872612e-05,
-      "loss": 0.4463,
+      "grad_norm": 3.547678232192993,
+      "learning_rate": 1.4171974522292993e-05,
+      "loss": 0.2737,
       "step": 540
     },
     {
       "epoch": 1.78,
-      "grad_norm": 4.211310863494873,
-      "learning_rate": 1.1305732484076434e-05,
-      "loss": 0.4229,
+      "grad_norm": 5.212786674499512,
+      "learning_rate": 1.0987261146496815e-05,
+      "loss": 0.2503,
       "step": 560
     },
     {
       "epoch": 1.85,
-      "grad_norm": 3.5312132835388184,
-      "learning_rate": 8.121019108280255e-06,
-      "loss": 0.4013,
+      "grad_norm": 3.4162983894348145,
+      "learning_rate": 7.802547770700637e-06,
+      "loss": 0.253,
       "step": 580
     },
     {
       "epoch": 1.91,
-      "grad_norm": 4.966080665588379,
-      "learning_rate": 4.936305732484077e-06,
-      "loss": 0.4346,
+      "grad_norm": 5.042104244232178,
+      "learning_rate": 4.6178343949044585e-06,
+      "loss": 0.2425,
       "step": 600
     },
     {
       "epoch": 1.91,
-      "eval_accuracy": 0.8066666666666666,
-      "eval_loss": 0.6526193618774414,
-      "eval_runtime": 14.3611,
-      "eval_samples_per_second": 73.114,
-      "eval_steps_per_second": 9.192,
+      "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": 2.851719856262207,
-      "learning_rate": 1.7515923566878982e-06,
-      "loss": 0.4156,
+      "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": 0.7493580146959633,
-      "train_runtime": 668.3994,
-      "train_samples_per_second": 30.027,
-      "train_steps_per_second": 0.94
+      "train_loss": 0.33379530375170857,
+      "train_runtime": 590.5675,
+      "train_samples_per_second": 33.984,
+      "train_steps_per_second": 1.063
     }
   ],
   "logging_steps": 20,