Training in progress, step 100

README.md

@@ -1,10 +1,11 @@
 ---
 license: apache-2.0
+base_model: google/vit-base-patch16-224-in21k
 tags:
 - image-classification
 - generated_from_trainer
 datasets:
-- beans
+- renovation
 metrics:
 - accuracy
 model-index:
@@ -15,14 +16,14 @@ model-index:
       type: image-classification
     dataset:
       name: beans
-      type: beans
+      type: renovation
       config: default
       split: validation
       args: default
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.9924812030075187
+      value: 0.6575342465753424
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +33,8 @@ 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: 0.0290
-- Accuracy: 0.9925
+- Loss: 0.7925
+- Accuracy: 0.6575
 
 ## Model description
 
@@ -59,18 +60,21 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 4
+- mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.1265        | 1.54  | 100  | 0.0826          | 0.9774   |
-| 0.0417        | 3.08  | 200  | 0.0290          | 0.9925   |
+| 1.1711        | 0.81  | 100  | 1.0255          | 0.5982   |
+| 0.7083        | 1.61  | 200  | 0.7925          | 0.6575   |
+| 0.2479        | 2.42  | 300  | 0.8712          | 0.6941   |
+| 0.127         | 3.23  | 400  | 0.8440          | 0.6941   |
 
 
 ### Framework versions
 
-- Transformers 4.30.2
-- Pytorch 2.0.1+cu118
-- Datasets 2.13.1
-- Tokenizers 0.13.3
+- Transformers 4.39.1
+- Pytorch 2.2.1+cu121
+- Datasets 2.18.0
+- Tokenizers 0.15.2

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.9924812030075187,
-    "eval_loss": 0.029002541676163673,
-    "eval_runtime": 3.0711,
-    "eval_samples_per_second": 43.308,
-    "eval_steps_per_second": 5.536,
-    "total_flos": 3.205097416476426e+17,
-    "train_loss": 0.128345954647431,
-    "train_runtime": 213.2669,
-    "train_samples_per_second": 19.394,
-    "train_steps_per_second": 1.219
+    "eval_accuracy": 0.6575342465753424,
+    "eval_loss": 0.7924718260765076,
+    "eval_runtime": 8.3551,
+    "eval_samples_per_second": 26.211,
+    "eval_steps_per_second": 3.351,
+    "total_flos": 6.10974224738132e+17,
+    "train_loss": 0.575864625193419,
+    "train_runtime": 402.9871,
+    "train_samples_per_second": 19.564,
+    "train_steps_per_second": 1.231
 }

config.json

@@ -9,17 +9,25 @@
   "hidden_dropout_prob": 0.0,
   "hidden_size": 768,
   "id2label": {
-    "0": "angular_leaf_spot",
-    "1": "bean_rust",
-    "2": "healthy"
+    "0": "Not Applicable",
+    "1": "Very Poor",
+    "2": "Poor",
+    "3": "Fair",
+    "4": "Good",
+    "5": "Excellent",
+    "6": "Exceptional"
   },
   "image_size": 224,
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
-    "angular_leaf_spot": "0",
-    "bean_rust": "1",
-    "healthy": "2"
+    "Excellent": "5",
+    "Exceptional": "6",
+    "Fair": "3",
+    "Good": "4",
+    "Not Applicable": "0",
+    "Poor": "2",
+    "Very Poor": "1"
   },
   "layer_norm_eps": 1e-12,
   "model_type": "vit",
@@ -30,5 +38,5 @@
   "problem_type": "single_label_classification",
   "qkv_bias": true,
   "torch_dtype": "float32",
-  "transformers_version": "4.30.2"
+  "transformers_version": "4.39.1"
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
-    "eval_accuracy": 0.9924812030075187,
-    "eval_loss": 0.029002541676163673,
-    "eval_runtime": 3.0711,
-    "eval_samples_per_second": 43.308,
-    "eval_steps_per_second": 5.536
+    "eval_accuracy": 0.6575342465753424,
+    "eval_loss": 0.7924718260765076,
+    "eval_runtime": 8.3551,
+    "eval_samples_per_second": 26.211,
+    "eval_steps_per_second": 3.351
 }

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bc0e701d02c681d64cf03f95b2e4e3c64596003ffa835c199c1a4c94f4e600d
+size 343239356

preprocessor_config.json

@@ -1,4 +1,18 @@
 {
+  "_valid_processor_keys": [
+    "images",
+    "do_resize",
+    "size",
+    "resample",
+    "do_rescale",
+    "rescale_factor",
+    "do_normalize",
+    "image_mean",
+    "image_std",
+    "return_tensors",
+    "data_format",
+    "input_data_format"
+  ],
   "do_normalize": true,
   "do_rescale": true,
   "do_resize": true,

runs/Mar23_00-59-58_3ec072334407/events.out.tfevents.1711155621.3ec072334407.4424.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6939c36490543f577d179ea3c5f35918846f62077f12970a94f89611544a085
+size 16756

runs/Mar23_00-59-58_3ec072334407/events.out.tfevents.1711156033.3ec072334407.4424.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38452b8c1a99812a91c95bd04cd995d6bee735f5a483ea63e7c60bfa5d5e7c1e
+size 411

runs/Mar23_01-10-51_3ec072334407/events.out.tfevents.1711156269.3ec072334407.4424.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:617c092ccfcd3d59faac86ff8d7faf1f4b2597d874a29a333c88a2e19e80567b
+size 7305

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 4.0,
-    "total_flos": 3.205097416476426e+17,
-    "train_loss": 0.128345954647431,
-    "train_runtime": 213.2669,
-    "train_samples_per_second": 19.394,
-    "train_steps_per_second": 1.219
+    "total_flos": 6.10974224738132e+17,
+    "train_loss": 0.575864625193419,
+    "train_runtime": 402.9871,
+    "train_samples_per_second": 19.564,
+    "train_steps_per_second": 1.231
 }

trainer_state.json

@@ -1,199 +1,409 @@
 {
-  "best_metric": 0.029002541676163673,
+  "best_metric": 0.7924718260765076,
   "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-200",
   "epoch": 4.0,
-  "global_step": 260,
+  "eval_steps": 100,
+  "global_step": 496,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.15,
-      "learning_rate": 0.00019230769230769233,
-      "loss": 0.8284,
+      "epoch": 0.08,
+      "grad_norm": 2.0731146335601807,
+      "learning_rate": 0.00019596774193548388,
+      "loss": 1.5546,
       "step": 10
     },
     {
-      "epoch": 0.31,
-      "learning_rate": 0.00018461538461538463,
-      "loss": 0.2956,
+      "epoch": 0.16,
+      "grad_norm": 1.9955084323883057,
+      "learning_rate": 0.00019193548387096775,
+      "loss": 1.3431,
       "step": 20
     },
     {
-      "epoch": 0.46,
-      "learning_rate": 0.00017692307692307693,
-      "loss": 0.3667,
+      "epoch": 0.24,
+      "grad_norm": 1.4386166334152222,
+      "learning_rate": 0.00018790322580645164,
+      "loss": 1.1702,
       "step": 30
     },
     {
-      "epoch": 0.62,
-      "learning_rate": 0.00016923076923076923,
-      "loss": 0.2713,
+      "epoch": 0.32,
+      "grad_norm": 2.1243135929107666,
+      "learning_rate": 0.00018387096774193548,
+      "loss": 1.1703,
       "step": 40
     },
     {
-      "epoch": 0.77,
-      "learning_rate": 0.00016153846153846155,
-      "loss": 0.2125,
+      "epoch": 0.4,
+      "grad_norm": 1.4760087728500366,
+      "learning_rate": 0.00017983870967741935,
+      "loss": 1.0635,
       "step": 50
     },
     {
-      "epoch": 0.92,
-      "learning_rate": 0.00015384615384615385,
-      "loss": 0.1852,
+      "epoch": 0.48,
+      "grad_norm": 1.3573176860809326,
+      "learning_rate": 0.00017580645161290325,
+      "loss": 1.0524,
       "step": 60
     },
     {
-      "epoch": 1.08,
-      "learning_rate": 0.00014615384615384615,
-      "loss": 0.1254,
+      "epoch": 0.56,
+      "grad_norm": 3.1192474365234375,
+      "learning_rate": 0.00017177419354838711,
+      "loss": 1.0872,
       "step": 70
     },
     {
-      "epoch": 1.23,
-      "learning_rate": 0.00013923076923076923,
-      "loss": 0.1552,
+      "epoch": 0.65,
+      "grad_norm": 2.4728469848632812,
+      "learning_rate": 0.00016774193548387098,
+      "loss": 0.87,
       "step": 80
     },
     {
-      "epoch": 1.38,
-      "learning_rate": 0.00013153846153846156,
-      "loss": 0.0474,
+      "epoch": 0.73,
+      "grad_norm": 3.819819688796997,
+      "learning_rate": 0.00016370967741935485,
+      "loss": 1.2173,
       "step": 90
     },
     {
-      "epoch": 1.54,
-      "learning_rate": 0.00012384615384615385,
-      "loss": 0.1265,
+      "epoch": 0.81,
+      "grad_norm": 1.585181713104248,
+      "learning_rate": 0.00015967741935483872,
+      "loss": 1.1711,
       "step": 100
     },
     {
-      "epoch": 1.54,
-      "eval_accuracy": 0.9774436090225563,
-      "eval_loss": 0.08259306848049164,
-      "eval_runtime": 2.625,
-      "eval_samples_per_second": 50.666,
-      "eval_steps_per_second": 6.476,
+      "epoch": 0.81,
+      "eval_accuracy": 0.5981735159817352,
+      "eval_loss": 1.0254943370819092,
+      "eval_runtime": 7.7395,
+      "eval_samples_per_second": 28.296,
+      "eval_steps_per_second": 3.618,
       "step": 100
     },
     {
-      "epoch": 1.69,
-      "learning_rate": 0.00011615384615384617,
-      "loss": 0.1154,
+      "epoch": 0.89,
+      "grad_norm": 1.5606015920639038,
+      "learning_rate": 0.0001556451612903226,
+      "loss": 0.9609,
       "step": 110
     },
     {
-      "epoch": 1.85,
-      "learning_rate": 0.00010846153846153846,
-      "loss": 0.1548,
+      "epoch": 0.97,
+      "grad_norm": 2.0005290508270264,
+      "learning_rate": 0.00015161290322580646,
+      "loss": 0.8673,
       "step": 120
     },
     {
-      "epoch": 2.0,
-      "learning_rate": 0.00010076923076923077,
-      "loss": 0.0739,
+      "epoch": 1.05,
+      "grad_norm": 1.7934025526046753,
+      "learning_rate": 0.00014758064516129032,
+      "loss": 0.7807,
       "step": 130
     },
     {
-      "epoch": 2.15,
-      "learning_rate": 9.307692307692309e-05,
-      "loss": 0.0507,
+      "epoch": 1.13,
+      "grad_norm": 2.0995261669158936,
+      "learning_rate": 0.00014354838709677422,
+      "loss": 0.7211,
       "step": 140
     },
     {
-      "epoch": 2.31,
-      "learning_rate": 8.538461538461538e-05,
-      "loss": 0.0479,
+      "epoch": 1.21,
+      "grad_norm": 1.2384763956069946,
+      "learning_rate": 0.0001395161290322581,
+      "loss": 0.8051,
       "step": 150
     },
     {
-      "epoch": 2.46,
-      "learning_rate": 7.76923076923077e-05,
-      "loss": 0.0475,
+      "epoch": 1.29,
+      "grad_norm": 3.001422643661499,
+      "learning_rate": 0.00013548387096774193,
+      "loss": 0.9052,
       "step": 160
     },
     {
-      "epoch": 2.62,
-      "learning_rate": 7e-05,
-      "loss": 0.0367,
+      "epoch": 1.37,
+      "grad_norm": 1.8868207931518555,
+      "learning_rate": 0.0001314516129032258,
+      "loss": 0.7217,
       "step": 170
     },
     {
-      "epoch": 2.77,
-      "learning_rate": 6.23076923076923e-05,
-      "loss": 0.0648,
+      "epoch": 1.45,
+      "grad_norm": 1.7666105031967163,
+      "learning_rate": 0.0001274193548387097,
+      "loss": 0.728,
       "step": 180
     },
     {
-      "epoch": 2.92,
-      "learning_rate": 5.461538461538461e-05,
-      "loss": 0.0201,
+      "epoch": 1.53,
+      "grad_norm": 2.378178358078003,
+      "learning_rate": 0.00012338709677419356,
+      "loss": 0.7242,
       "step": 190
     },
     {
-      "epoch": 3.08,
-      "learning_rate": 4.692307692307693e-05,
-      "loss": 0.0417,
+      "epoch": 1.61,
+      "grad_norm": 2.4882402420043945,
+      "learning_rate": 0.00011935483870967743,
+      "loss": 0.7083,
       "step": 200
     },
     {
-      "epoch": 3.08,
-      "eval_accuracy": 0.9924812030075187,
-      "eval_loss": 0.029002541676163673,
-      "eval_runtime": 2.6469,
-      "eval_samples_per_second": 50.247,
-      "eval_steps_per_second": 6.423,
+      "epoch": 1.61,
+      "eval_accuracy": 0.6575342465753424,
+      "eval_loss": 0.7924718260765076,
+      "eval_runtime": 7.9712,
+      "eval_samples_per_second": 27.474,
+      "eval_steps_per_second": 3.513,
       "step": 200
     },
     {
-      "epoch": 3.23,
-      "learning_rate": 3.923076923076923e-05,
-      "loss": 0.0126,
+      "epoch": 1.69,
+      "grad_norm": 3.762314796447754,
+      "learning_rate": 0.00011532258064516131,
+      "loss": 0.6794,
       "step": 210
     },
     {
-      "epoch": 3.38,
-      "learning_rate": 3.153846153846154e-05,
-      "loss": 0.0117,
+      "epoch": 1.77,
+      "grad_norm": 1.8527878522872925,
+      "learning_rate": 0.00011129032258064515,
+      "loss": 0.5424,
       "step": 220
     },
     {
-      "epoch": 3.54,
-      "learning_rate": 2.384615384615385e-05,
-      "loss": 0.0109,
+      "epoch": 1.85,
+      "grad_norm": 2.1447601318359375,
+      "learning_rate": 0.00010725806451612903,
+      "loss": 0.7186,
       "step": 230
     },
     {
-      "epoch": 3.69,
-      "learning_rate": 1.6153846153846154e-05,
-      "loss": 0.0123,
+      "epoch": 1.94,
+      "grad_norm": 2.2273943424224854,
+      "learning_rate": 0.0001032258064516129,
+      "loss": 0.7786,
       "step": 240
     },
     {
-      "epoch": 3.85,
-      "learning_rate": 8.461538461538462e-06,
-      "loss": 0.0113,
+      "epoch": 2.02,
+      "grad_norm": 1.1381219625473022,
+      "learning_rate": 9.919354838709678e-05,
+      "loss": 0.5616,
       "step": 250
     },
     {
-      "epoch": 4.0,
-      "learning_rate": 7.692307692307694e-07,
-      "loss": 0.0106,
+      "epoch": 2.1,
+      "grad_norm": 1.41716730594635,
+      "learning_rate": 9.516129032258065e-05,
+      "loss": 0.3615,
       "step": 260
     },
+    {
+      "epoch": 2.18,
+      "grad_norm": 1.193400263786316,
+      "learning_rate": 9.112903225806452e-05,
+      "loss": 0.3589,
+      "step": 270
+    },
+    {
+      "epoch": 2.26,
+      "grad_norm": 1.1476421356201172,
+      "learning_rate": 8.709677419354839e-05,
+      "loss": 0.302,
+      "step": 280
+    },
+    {
+      "epoch": 2.34,
+      "grad_norm": 2.02689790725708,
+      "learning_rate": 8.306451612903227e-05,
+      "loss": 0.4161,
+      "step": 290
+    },
+    {
+      "epoch": 2.42,
+      "grad_norm": 1.2040166854858398,
+      "learning_rate": 7.903225806451613e-05,
+      "loss": 0.2479,
+      "step": 300
+    },
+    {
+      "epoch": 2.42,
+      "eval_accuracy": 0.6940639269406392,
+      "eval_loss": 0.871184766292572,
+      "eval_runtime": 7.1874,
+      "eval_samples_per_second": 30.47,
+      "eval_steps_per_second": 3.896,
+      "step": 300
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 4.465162754058838,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 0.2949,
+      "step": 310
+    },
+    {
+      "epoch": 2.58,
+      "grad_norm": 2.3733272552490234,
+      "learning_rate": 7.096774193548388e-05,
+      "loss": 0.2599,
+      "step": 320
+    },
+    {
+      "epoch": 2.66,
+      "grad_norm": 3.527358293533325,
+      "learning_rate": 6.693548387096774e-05,
+      "loss": 0.4337,
+      "step": 330
+    },
+    {
+      "epoch": 2.74,
+      "grad_norm": 2.8637776374816895,
+      "learning_rate": 6.290322580645161e-05,
+      "loss": 0.3137,
+      "step": 340
+    },
+    {
+      "epoch": 2.82,
+      "grad_norm": 4.198659896850586,
+      "learning_rate": 5.887096774193549e-05,
+      "loss": 0.3192,
+      "step": 350
+    },
+    {
+      "epoch": 2.9,
+      "grad_norm": 4.358771324157715,
+      "learning_rate": 5.4838709677419355e-05,
+      "loss": 0.3796,
+      "step": 360
+    },
+    {
+      "epoch": 2.98,
+      "grad_norm": 3.1412558555603027,
+      "learning_rate": 5.080645161290323e-05,
+      "loss": 0.2622,
+      "step": 370
+    },
+    {
+      "epoch": 3.06,
+      "grad_norm": 1.4319863319396973,
+      "learning_rate": 4.67741935483871e-05,
+      "loss": 0.2016,
+      "step": 380
+    },
+    {
+      "epoch": 3.15,
+      "grad_norm": 0.5633509755134583,
+      "learning_rate": 4.2741935483870973e-05,
+      "loss": 0.1276,
+      "step": 390
+    },
+    {
+      "epoch": 3.23,
+      "grad_norm": 0.31856054067611694,
+      "learning_rate": 3.870967741935484e-05,
+      "loss": 0.127,
+      "step": 400
+    },
+    {
+      "epoch": 3.23,
+      "eval_accuracy": 0.6940639269406392,
+      "eval_loss": 0.8440136313438416,
+      "eval_runtime": 7.6031,
+      "eval_samples_per_second": 28.804,
+      "eval_steps_per_second": 3.683,
+      "step": 400
+    },
+    {
+      "epoch": 3.31,
+      "grad_norm": 0.2085038274526596,
+      "learning_rate": 3.467741935483872e-05,
+      "loss": 0.1289,
+      "step": 410
+    },
+    {
+      "epoch": 3.39,
+      "grad_norm": 2.471668004989624,
+      "learning_rate": 3.0645161290322585e-05,
+      "loss": 0.1495,
+      "step": 420
+    },
+    {
+      "epoch": 3.47,
+      "grad_norm": 1.026688575744629,
+      "learning_rate": 2.661290322580645e-05,
+      "loss": 0.1389,
+      "step": 430
+    },
+    {
+      "epoch": 3.55,
+      "grad_norm": 4.408120155334473,
+      "learning_rate": 2.258064516129032e-05,
+      "loss": 0.1799,
+      "step": 440
+    },
+    {
+      "epoch": 3.63,
+      "grad_norm": 1.7386049032211304,
+      "learning_rate": 1.8548387096774193e-05,
+      "loss": 0.0727,
+      "step": 450
+    },
+    {
+      "epoch": 3.71,
+      "grad_norm": 0.9766908884048462,
+      "learning_rate": 1.4516129032258066e-05,
+      "loss": 0.1062,
+      "step": 460
+    },
+    {
+      "epoch": 3.79,
+      "grad_norm": 0.2413896918296814,
+      "learning_rate": 1.0483870967741936e-05,
+      "loss": 0.1303,
+      "step": 470
+    },
+    {
+      "epoch": 3.87,
+      "grad_norm": 1.249718427658081,
+      "learning_rate": 6.451612903225806e-06,
+      "loss": 0.1025,
+      "step": 480
+    },
+    {
+      "epoch": 3.95,
+      "grad_norm": 3.0604677200317383,
+      "learning_rate": 2.4193548387096776e-06,
+      "loss": 0.1515,
+      "step": 490
+    },
     {
       "epoch": 4.0,
-      "step": 260,
-      "total_flos": 3.205097416476426e+17,
-      "train_loss": 0.128345954647431,
-      "train_runtime": 213.2669,
-      "train_samples_per_second": 19.394,
-      "train_steps_per_second": 1.219
+      "step": 496,
+      "total_flos": 6.10974224738132e+17,
+      "train_loss": 0.575864625193419,
+      "train_runtime": 402.9871,
+      "train_samples_per_second": 19.564,
+      "train_steps_per_second": 1.231
     }
   ],
-  "max_steps": 260,
+  "logging_steps": 10,
+  "max_steps": 496,
+  "num_input_tokens_seen": 0,
   "num_train_epochs": 4,
-  "total_flos": 3.205097416476426e+17,
+  "save_steps": 100,
+  "total_flos": 6.10974224738132e+17,
+  "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:afc80ecabdb2ed38d16d114ee107514b31aee475b41ef1ccec775f3ebc1eff49
-size 3963
+oid sha256:fc8fb6c8455921fc914ef6cdd8d22a85ffa860ed7bbef046a7bd772a28f8da24
+size 4920