🍻 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
 model-index:
 - name: ryan_model314_3
@@ -13,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # ryan_model314_3
 
-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 None 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 beans dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.2696
 - Na Accuracy: 0.96

all_results.json

@@ -1,13 +1,14 @@
 {
-    "epoch": 0.51,
-    "eval_loss": 0.2651739716529846,
-    "eval_na_accuracy": 0.9371701432520734,
-    "eval_ordinal_mae": 1.021151511080242,
-    "eval_runtime": 162.3376,
-    "eval_samples_per_second": 24.511,
-    "eval_steps_per_second": 3.068,
-    "train_loss": 0.29898834055120294,
-    "train_runtime": 4048.1416,
-    "train_samples_per_second": 34.064,
-    "train_steps_per_second": 2.129
+    "epoch": 2.0,
+    "eval_loss": 0.2696247398853302,
+    "eval_na_accuracy": 0.96,
+    "eval_ordinal_accuracy": 0.5,
+    "eval_ordinal_mae": 1.168388410445914,
+    "eval_runtime": 4.4189,
+    "eval_samples_per_second": 22.63,
+    "eval_steps_per_second": 2.942,
+    "train_loss": 0.3214722058129689,
+    "train_runtime": 262.7567,
+    "train_samples_per_second": 7.612,
+    "train_steps_per_second": 0.48
 }

eval_results.json

@@ -1,9 +1,10 @@
 {
-    "epoch": 0.51,
-    "eval_loss": 0.2651739716529846,
-    "eval_na_accuracy": 0.9371701432520734,
-    "eval_ordinal_mae": 1.021151511080242,
-    "eval_runtime": 162.3376,
-    "eval_samples_per_second": 24.511,
-    "eval_steps_per_second": 3.068
+    "epoch": 2.0,
+    "eval_loss": 0.2696247398853302,
+    "eval_na_accuracy": 0.96,
+    "eval_ordinal_accuracy": 0.5,
+    "eval_ordinal_mae": 1.168388410445914,
+    "eval_runtime": 4.4189,
+    "eval_samples_per_second": 22.63,
+    "eval_steps_per_second": 2.942
 }

runs/Mar27_07-38-15_ryanserver/events.out.tfevents.1711539836.ryanserver.4146.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f857d222c6c7045f0baeea737e138c33ca42010ed029a92f0da9178649dc6d9c
+size 521

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 0.51,
-    "train_loss": 0.29898834055120294,
-    "train_runtime": 4048.1416,
-    "train_samples_per_second": 34.064,
-    "train_steps_per_second": 2.129
+    "epoch": 2.0,
+    "train_loss": 0.3214722058129689,
+    "train_runtime": 262.7567,
+    "train_samples_per_second": 7.612,
+    "train_steps_per_second": 0.48
 }

trainer_state.json

@@ -1,216 +1,65 @@
 {
-  "best_metric": 0.2651739716529846,
-  "best_model_checkpoint": "./ryan_model314_3/checkpoint-1000",
-  "epoch": 0.5104408352668214,
-  "eval_steps": 100,
-  "global_step": 1100,
+  "best_metric": 0.2696247398853302,
+  "best_model_checkpoint": "./ryan_model314_3/checkpoint-100",
+  "epoch": 2.0,
+  "eval_steps": 50,
+  "global_step": 126,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.05,
-      "grad_norm": 1.358393669128418,
-      "learning_rate": 0.00019767981438515082,
-      "loss": 0.3676,
-      "step": 100
-    },
-    {
-      "epoch": 0.05,
-      "eval_loss": 0.34232085943222046,
-      "eval_na_accuracy": 0.9273686855993968,
-      "eval_ordinal_mae": 1.1293019706063574,
-      "eval_runtime": 381.3744,
-      "eval_samples_per_second": 10.433,
-      "eval_steps_per_second": 1.306,
-      "step": 100
-    },
-    {
-      "epoch": 0.09,
-      "grad_norm": 0.9884235262870789,
-      "learning_rate": 0.00019535962877030162,
-      "loss": 0.3329,
-      "step": 200
-    },
-    {
-      "epoch": 0.09,
-      "eval_loss": 0.31356191635131836,
-      "eval_na_accuracy": 0.9313897964312642,
-      "eval_ordinal_mae": 1.070580373245832,
-      "eval_runtime": 164.3074,
-      "eval_samples_per_second": 24.217,
-      "eval_steps_per_second": 3.031,
-      "step": 200
-    },
-    {
-      "epoch": 0.14,
-      "grad_norm": 1.2422517538070679,
-      "learning_rate": 0.00019303944315545243,
-      "loss": 0.3134,
-      "step": 300
-    },
-    {
-      "epoch": 0.14,
-      "eval_loss": 0.3302006423473358,
-      "eval_na_accuracy": 0.9165619502387534,
-      "eval_ordinal_mae": 1.1219553900231423,
-      "eval_runtime": 164.3869,
-      "eval_samples_per_second": 24.205,
-      "eval_steps_per_second": 3.029,
-      "step": 300
-    },
-    {
-      "epoch": 0.19,
-      "grad_norm": 1.1815475225448608,
-      "learning_rate": 0.00019071925754060324,
-      "loss": 0.314,
-      "step": 400
-    },
-    {
-      "epoch": 0.19,
-      "eval_loss": 0.2992381155490875,
-      "eval_na_accuracy": 0.9256094496104549,
-      "eval_ordinal_mae": 0.820222080006495,
-      "eval_runtime": 165.1522,
-      "eval_samples_per_second": 24.093,
-      "eval_steps_per_second": 3.015,
-      "step": 400
-    },
-    {
-      "epoch": 0.23,
-      "grad_norm": 0.47366032004356384,
-      "learning_rate": 0.00018839907192575407,
-      "loss": 0.2965,
-      "step": 500
-    },
-    {
-      "epoch": 0.23,
-      "eval_loss": 0.319803923368454,
-      "eval_na_accuracy": 0.9248554913294798,
-      "eval_ordinal_mae": 1.2210362517446196,
-      "eval_runtime": 163.23,
-      "eval_samples_per_second": 24.377,
-      "eval_steps_per_second": 3.051,
-      "step": 500
+      "epoch": 0.79,
+      "grad_norm": 0.766051709651947,
+      "learning_rate": 0.00012063492063492063,
+      "loss": 0.4088,
+      "step": 50
     },
     {
-      "epoch": 0.28,
-      "grad_norm": 1.8707951307296753,
-      "learning_rate": 0.00018607888631090488,
-      "loss": 0.3068,
-      "step": 600
+      "epoch": 0.79,
+      "eval_loss": 0.30989590287208557,
+      "eval_na_accuracy": 0.93,
+      "eval_ordinal_accuracy": 0.43478260869565216,
+      "eval_ordinal_mae": 1.1052913616381967,
+      "eval_runtime": 14.2973,
+      "eval_samples_per_second": 6.994,
+      "eval_steps_per_second": 0.909,
+      "step": 50
     },
     {
-      "epoch": 0.28,
-      "eval_loss": 0.26730772852897644,
-      "eval_na_accuracy": 0.9371701432520734,
-      "eval_ordinal_mae": 1.103637127990878,
-      "eval_runtime": 160.1368,
-      "eval_samples_per_second": 24.848,
-      "eval_steps_per_second": 3.11,
-      "step": 600
-    },
-    {
-      "epoch": 0.32,
-      "grad_norm": 1.1020443439483643,
-      "learning_rate": 0.0001837587006960557,
-      "loss": 0.2824,
-      "step": 700
-    },
-    {
-      "epoch": 0.32,
-      "eval_loss": 0.2921823561191559,
-      "eval_na_accuracy": 0.9371701432520734,
-      "eval_ordinal_mae": 1.4977284913306896,
-      "eval_runtime": 163.7628,
-      "eval_samples_per_second": 24.297,
-      "eval_steps_per_second": 3.041,
-      "step": 700
-    },
-    {
-      "epoch": 0.37,
-      "grad_norm": 1.2855117321014404,
-      "learning_rate": 0.0001814385150812065,
-      "loss": 0.2914,
-      "step": 800
-    },
-    {
-      "epoch": 0.37,
-      "eval_loss": 0.2798122763633728,
-      "eval_na_accuracy": 0.9384267403870319,
-      "eval_ordinal_mae": 0.7788859930601368,
-      "eval_runtime": 163.524,
-      "eval_samples_per_second": 24.333,
-      "eval_steps_per_second": 3.045,
-      "step": 800
-    },
-    {
-      "epoch": 0.42,
-      "grad_norm": 1.4634666442871094,
-      "learning_rate": 0.00017911832946635733,
-      "loss": 0.2968,
-      "step": 900
-    },
-    {
-      "epoch": 0.42,
-      "eval_loss": 0.27096793055534363,
-      "eval_na_accuracy": 0.9369188238250816,
-      "eval_ordinal_mae": 0.9694435305190251,
-      "eval_runtime": 163.2761,
-      "eval_samples_per_second": 24.37,
-      "eval_steps_per_second": 3.05,
-      "step": 900
-    },
-    {
-      "epoch": 0.46,
-      "grad_norm": 0.3547471761703491,
-      "learning_rate": 0.00017679814385150814,
-      "loss": 0.2433,
-      "step": 1000
-    },
-    {
-      "epoch": 0.46,
-      "eval_loss": 0.2651739716529846,
-      "eval_na_accuracy": 0.9371701432520734,
-      "eval_ordinal_mae": 1.021151511080242,
-      "eval_runtime": 165.2163,
-      "eval_samples_per_second": 24.084,
-      "eval_steps_per_second": 3.014,
-      "step": 1000
-    },
-    {
-      "epoch": 0.51,
-      "grad_norm": 0.6242479085922241,
-      "learning_rate": 0.00017447795823665894,
-      "loss": 0.2438,
-      "step": 1100
+      "epoch": 1.59,
+      "grad_norm": 0.8156992793083191,
+      "learning_rate": 4.126984126984127e-05,
+      "loss": 0.2813,
+      "step": 100
     },
     {
-      "epoch": 0.51,
-      "eval_loss": 0.2782880365848541,
-      "eval_na_accuracy": 0.9389293792410154,
-      "eval_ordinal_mae": 0.8154305260353155,
-      "eval_runtime": 161.9425,
-      "eval_samples_per_second": 24.57,
-      "eval_steps_per_second": 3.075,
-      "step": 1100
+      "epoch": 1.59,
+      "eval_loss": 0.2696247398853302,
+      "eval_na_accuracy": 0.96,
+      "eval_ordinal_accuracy": 0.5,
+      "eval_ordinal_mae": 1.168388410445914,
+      "eval_runtime": 4.0273,
+      "eval_samples_per_second": 24.831,
+      "eval_steps_per_second": 3.228,
+      "step": 100
     },
     {
-      "epoch": 0.51,
-      "step": 1100,
-      "total_flos": 1.363907912564736e+18,
-      "train_loss": 0.29898834055120294,
-      "train_runtime": 4048.1416,
-      "train_samples_per_second": 34.064,
-      "train_steps_per_second": 2.129
+      "epoch": 2.0,
+      "step": 126,
+      "total_flos": 1.5498953551872e+17,
+      "train_loss": 0.3214722058129689,
+      "train_runtime": 262.7567,
+      "train_samples_per_second": 7.612,
+      "train_steps_per_second": 0.48
     }
   ],
-  "logging_steps": 100,
-  "max_steps": 8620,
+  "logging_steps": 50,
+  "max_steps": 126,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 4,
-  "save_steps": 100,
-  "total_flos": 1.363907912564736e+18,
+  "num_train_epochs": 2,
+  "save_steps": 50,
+  "total_flos": 1.5498953551872e+17,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null