m0vie

README.md

@@ -19,8 +19,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 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 beans dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1201
-- Accuracy: 0.9688
+- Loss: 0.1760
+- Accuracy: 0.9531
 
 ## Model description
 
@@ -52,8 +52,8 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss | Accuracy |
 |:-------------:|:------:|:----:|:---------------:|:--------:|
-| 0.0846        | 1.5385 | 100  | 0.0614          | 0.9774   |
-| 0.0121        | 3.0769 | 200  | 0.0188          | 1.0      |
+| 0.0725        | 1.5385 | 100  | 0.1634          | 0.9474   |
+| 0.0145        | 3.0769 | 200  | 0.0318          | 0.9925   |
 
 
 ### Framework versions

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.96875,
-    "eval_loss": 0.12014883756637573,
-    "eval_runtime": 1.4118,
-    "eval_samples_per_second": 90.668,
-    "eval_steps_per_second": 11.333,
+    "eval_accuracy": 0.953125,
+    "eval_loss": 0.1760333776473999,
+    "eval_runtime": 1.8596,
+    "eval_samples_per_second": 68.831,
+    "eval_steps_per_second": 8.604,
     "total_flos": 3.205097416476426e+17,
-    "train_loss": 0.1095780831690018,
-    "train_runtime": 90.7321,
-    "train_samples_per_second": 45.585,
-    "train_steps_per_second": 2.866
+    "train_loss": 0.12231931924246825,
+    "train_runtime": 89.2176,
+    "train_samples_per_second": 46.359,
+    "train_steps_per_second": 2.914
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f243f5eda82922461d9227c7e69fbef747dbd510a9019cb215e3dc6f4d2ca0d2
+oid sha256:49d38e18efa6bde3220d6e044b699c03976ae00fb42b00a89f4ace18bb9c207d
 size 343227052

runs/Jul08_06-47-03_33d8eb963be1/events.out.tfevents.1720422777.33d8eb963be1.1201.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba6822308c6b30b8d4c704d340dd13d2068efd3a878e5974dbe35f59ff7c894f
+size 11323

runs/Jul08_06-47-03_33d8eb963be1/events.out.tfevents.1720423002.33d8eb963be1.1201.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6824763ed256fa5d26941b629d6118b337b1235a937bc7fa5bb03792191abff
+size 411

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.96875,
-    "eval_loss": 0.12014883756637573,
-    "eval_runtime": 1.4118,
-    "eval_samples_per_second": 90.668,
-    "eval_steps_per_second": 11.333
+    "eval_accuracy": 0.953125,
+    "eval_loss": 0.1760333776473999,
+    "eval_runtime": 1.8596,
+    "eval_samples_per_second": 68.831,
+    "eval_steps_per_second": 8.604
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
     "total_flos": 3.205097416476426e+17,
-    "train_loss": 0.1095780831690018,
-    "train_runtime": 90.7321,
-    "train_samples_per_second": 45.585,
-    "train_steps_per_second": 2.866
+    "train_loss": 0.12231931924246825,
+    "train_runtime": 89.2176,
+    "train_samples_per_second": 46.359,
+    "train_steps_per_second": 2.914
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.018796762451529503,
+  "best_metric": 0.03179202973842621,
   "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-200",
   "epoch": 4.0,
   "eval_steps": 100,
@@ -10,212 +10,212 @@
   "log_history": [
     {
       "epoch": 0.15384615384615385,
-      "grad_norm": 1.4108372926712036,
+      "grad_norm": 1.3854262828826904,
       "learning_rate": 0.00019230769230769233,
-      "loss": 0.7924,
+      "loss": 0.8006,
       "step": 10
     },
     {
       "epoch": 0.3076923076923077,
-      "grad_norm": 2.044903039932251,
+      "grad_norm": 4.267582893371582,
       "learning_rate": 0.00018461538461538463,
-      "loss": 0.301,
+      "loss": 0.3319,
       "step": 20
     },
     {
       "epoch": 0.46153846153846156,
-      "grad_norm": 0.7039980888366699,
+      "grad_norm": 1.7108104228973389,
       "learning_rate": 0.00017692307692307693,
-      "loss": 0.1992,
+      "loss": 0.2789,
       "step": 30
     },
     {
       "epoch": 0.6153846153846154,
-      "grad_norm": 1.1257214546203613,
+      "grad_norm": 1.280118465423584,
       "learning_rate": 0.00016923076923076923,
-      "loss": 0.2634,
+      "loss": 0.2837,
       "step": 40
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 0.34912949800491333,
+      "grad_norm": 8.134176254272461,
       "learning_rate": 0.00016153846153846155,
-      "loss": 0.1518,
+      "loss": 0.2461,
       "step": 50
     },
     {
       "epoch": 0.9230769230769231,
-      "grad_norm": 4.554572582244873,
+      "grad_norm": 2.2806813716888428,
       "learning_rate": 0.00015384615384615385,
-      "loss": 0.23,
+      "loss": 0.2131,
       "step": 60
     },
     {
       "epoch": 1.0769230769230769,
-      "grad_norm": 0.4593968093395233,
+      "grad_norm": 0.2824852168560028,
       "learning_rate": 0.00014615384615384615,
-      "loss": 0.0491,
+      "loss": 0.1639,
       "step": 70
     },
     {
       "epoch": 1.2307692307692308,
-      "grad_norm": 0.13646918535232544,
+      "grad_norm": 0.39076101779937744,
       "learning_rate": 0.00013846153846153847,
-      "loss": 0.1,
+      "loss": 0.0845,
       "step": 80
     },
     {
       "epoch": 1.3846153846153846,
-      "grad_norm": 5.695180416107178,
+      "grad_norm": 0.1591070592403412,
       "learning_rate": 0.00013076923076923077,
-      "loss": 0.0552,
+      "loss": 0.0721,
       "step": 90
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 1.3406460285186768,
+      "grad_norm": 2.908008575439453,
       "learning_rate": 0.0001230769230769231,
-      "loss": 0.0846,
+      "loss": 0.0725,
       "step": 100
     },
     {
       "epoch": 1.5384615384615383,
-      "eval_accuracy": 0.9774436090225563,
-      "eval_loss": 0.061424773186445236,
-      "eval_runtime": 1.708,
-      "eval_samples_per_second": 77.869,
-      "eval_steps_per_second": 9.953,
+      "eval_accuracy": 0.9473684210526315,
+      "eval_loss": 0.1633957028388977,
+      "eval_runtime": 1.2488,
+      "eval_samples_per_second": 106.503,
+      "eval_steps_per_second": 13.613,
       "step": 100
     },
     {
       "epoch": 1.6923076923076923,
-      "grad_norm": 1.4606086015701294,
+      "grad_norm": 3.4353740215301514,
       "learning_rate": 0.00011538461538461538,
-      "loss": 0.1462,
+      "loss": 0.1748,
       "step": 110
     },
     {
       "epoch": 1.8461538461538463,
-      "grad_norm": 0.5630506873130798,
+      "grad_norm": 3.038853883743286,
       "learning_rate": 0.0001076923076923077,
-      "loss": 0.0893,
+      "loss": 0.1436,
       "step": 120
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.09152624756097794,
+      "grad_norm": 2.874983310699463,
       "learning_rate": 0.0001,
-      "loss": 0.0368,
+      "loss": 0.0415,
       "step": 130
     },
     {
       "epoch": 2.1538461538461537,
-      "grad_norm": 0.9141150116920471,
+      "grad_norm": 0.20417748391628265,
       "learning_rate": 9.230769230769232e-05,
-      "loss": 0.0192,
+      "loss": 0.0394,
       "step": 140
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 1.7744202613830566,
+      "grad_norm": 9.560956954956055,
       "learning_rate": 8.461538461538461e-05,
-      "loss": 0.0423,
+      "loss": 0.0438,
       "step": 150
     },
     {
       "epoch": 2.4615384615384617,
-      "grad_norm": 0.14094310998916626,
+      "grad_norm": 0.3993445634841919,
       "learning_rate": 7.692307692307693e-05,
-      "loss": 0.0894,
+      "loss": 0.0366,
       "step": 160
     },
     {
       "epoch": 2.6153846153846154,
-      "grad_norm": 0.2033337652683258,
+      "grad_norm": 0.07924646139144897,
       "learning_rate": 6.923076923076924e-05,
-      "loss": 0.0158,
+      "loss": 0.0157,
       "step": 170
     },
     {
       "epoch": 2.769230769230769,
-      "grad_norm": 2.038670778274536,
+      "grad_norm": 0.07146325707435608,
       "learning_rate": 6.153846153846155e-05,
-      "loss": 0.0682,
+      "loss": 0.0336,
       "step": 180
     },
     {
       "epoch": 2.9230769230769234,
-      "grad_norm": 0.5425238609313965,
+      "grad_norm": 1.8434661626815796,
       "learning_rate": 5.384615384615385e-05,
-      "loss": 0.0148,
+      "loss": 0.0176,
       "step": 190
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 0.12908649444580078,
+      "grad_norm": 0.07253334671258926,
       "learning_rate": 4.615384615384616e-05,
-      "loss": 0.0121,
+      "loss": 0.0145,
       "step": 200
     },
     {
       "epoch": 3.076923076923077,
-      "eval_accuracy": 1.0,
-      "eval_loss": 0.018796762451529503,
-      "eval_runtime": 2.1434,
-      "eval_samples_per_second": 62.05,
-      "eval_steps_per_second": 7.931,
+      "eval_accuracy": 0.9924812030075187,
+      "eval_loss": 0.03179202973842621,
+      "eval_runtime": 1.4947,
+      "eval_samples_per_second": 88.979,
+      "eval_steps_per_second": 11.373,
       "step": 200
     },
     {
       "epoch": 3.230769230769231,
-      "grad_norm": 0.07232276350259781,
+      "grad_norm": 0.0676412433385849,
       "learning_rate": 3.846153846153846e-05,
-      "loss": 0.0119,
+      "loss": 0.013,
       "step": 210
     },
     {
       "epoch": 3.3846153846153846,
-      "grad_norm": 0.05829685181379318,
+      "grad_norm": 0.06383011490106583,
       "learning_rate": 3.0769230769230774e-05,
-      "loss": 0.0131,
+      "loss": 0.0123,
       "step": 220
     },
     {
       "epoch": 3.5384615384615383,
-      "grad_norm": 0.06218134984374046,
+      "grad_norm": 0.06610537320375443,
       "learning_rate": 2.307692307692308e-05,
-      "loss": 0.0306,
+      "loss": 0.0119,
       "step": 230
     },
     {
       "epoch": 3.6923076923076925,
-      "grad_norm": 0.06792601197957993,
+      "grad_norm": 0.06182776391506195,
       "learning_rate": 1.5384615384615387e-05,
-      "loss": 0.011,
+      "loss": 0.0116,
       "step": 240
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 0.060047898441553116,
+      "grad_norm": 0.05951670929789543,
       "learning_rate": 7.692307692307694e-06,
-      "loss": 0.0108,
+      "loss": 0.0115,
       "step": 250
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.05698742717504501,
+      "grad_norm": 0.11315698176622391,
       "learning_rate": 0.0,
-      "loss": 0.0106,
+      "loss": 0.0117,
       "step": 260
     },
     {
       "epoch": 4.0,
       "step": 260,
       "total_flos": 3.205097416476426e+17,
-      "train_loss": 0.1095780831690018,
-      "train_runtime": 90.7321,
-      "train_samples_per_second": 45.585,
-      "train_steps_per_second": 2.866
+      "train_loss": 0.12231931924246825,
+      "train_runtime": 89.2176,
+      "train_samples_per_second": 46.359,
+      "train_steps_per_second": 2.914
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:79743f40a0fe1e7366d56248ce2933432324293fcaaef4908186b85c13cc78a3
+oid sha256:1068466c39135f47bf594575dd82b090510f428889e5c78115653611ddee98e8
 size 5112