Training in progress, epoch 1

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 3.0,
-    "eval_accuracy": 0.7727272727272727,
-    "eval_loss": 0.5769956111907959,
-    "eval_runtime": 0.3139,
-    "eval_samples_per_second": 70.075,
-    "eval_steps_per_second": 6.37,
-    "total_flos": 1.4093345754980352e+16,
-    "train_loss": 0.5933622784084744,
-    "train_runtime": 25.6802,
-    "train_samples_per_second": 22.079,
-    "train_steps_per_second": 0.35
+    "epoch": 5.0,
+    "eval_accuracy": 0.7272727272727273,
+    "eval_loss": 0.7899118065834045,
+    "eval_runtime": 0.2994,
+    "eval_samples_per_second": 73.482,
+    "eval_steps_per_second": 6.68,
+    "total_flos": 2.348890959163392e+16,
+    "train_loss": 0.2989959716796875,
+    "train_runtime": 40.2987,
+    "train_samples_per_second": 23.45,
+    "train_steps_per_second": 0.372
 }

config.json

@@ -1,18 +1,10 @@
 {
-  "_name_or_path": "swin-tiny-patch4-window7-224-finetuned-eurosat-kornia",
+  "_name_or_path": "google/vit-base-patch16-224",
   "architectures": [
-    "SwinForImageClassification"
+    "ViTForImageClassification"
   ],
   "attention_probs_dropout_prob": 0.0,
-  "depths": [
-    2,
-    2,
-    6,
-    2
-  ],
-  "drop_path_rate": 0.1,
-  "embed_dim": 96,
-  "encoder_stride": 32,
+  "encoder_stride": 16,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.0,
   "hidden_size": 768,
@@ -22,40 +14,19 @@
   },
   "image_size": 224,
   "initializer_range": 0.02,
+  "intermediate_size": 3072,
   "label2id": {
     "clethrionomys": 0,
     "sylvaemus": 1
   },
-  "layer_norm_eps": 1e-05,
-  "mlp_ratio": 4.0,
-  "model_type": "swin",
+  "layer_norm_eps": 1e-12,
+  "model_type": "vit",
+  "num_attention_heads": 12,
   "num_channels": 3,
-  "num_heads": [
-    3,
-    6,
-    12,
-    24
-  ],
-  "num_layers": 4,
-  "out_features": [
-    "stage4"
-  ],
-  "out_indices": [
-    4
-  ],
-  "patch_size": 4,
-  "path_norm": true,
+  "num_hidden_layers": 12,
+  "patch_size": 16,
   "problem_type": "single_label_classification",
   "qkv_bias": true,
-  "stage_names": [
-    "stem",
-    "stage1",
-    "stage2",
-    "stage3",
-    "stage4"
-  ],
   "torch_dtype": "float32",
-  "transformers_version": "4.35.2",
-  "use_absolute_embeddings": false,
-  "window_size": 7
+  "transformers_version": "4.35.2"
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "eval_accuracy": 0.7727272727272727,
-    "eval_loss": 0.5769956111907959,
-    "eval_runtime": 0.3139,
-    "eval_samples_per_second": 70.075,
-    "eval_steps_per_second": 6.37
+    "epoch": 5.0,
+    "eval_accuracy": 0.7272727272727273,
+    "eval_loss": 0.7899118065834045,
+    "eval_runtime": 0.2994,
+    "eval_samples_per_second": 73.482,
+    "eval_steps_per_second": 6.68
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec318a30c4de32991c1f68e1788d705fe3084cb7ac93b333eb218baf10fa3b8e
-size 110342832
+oid sha256:14017dea4a569ea98e9e99a06667d04fb040dc086df06f0b897a9222d1e5db17
+size 343223968

preprocessor_config.json

@@ -2,19 +2,18 @@
   "do_normalize": true,
   "do_rescale": true,
   "do_resize": true,
-  "feature_extractor_type": "ViTFeatureExtractor",
   "image_mean": [
-    0.485,
-    0.456,
-    0.406
+    0.5,
+    0.5,
+    0.5
   ],
-  "image_processor_type": "ViTFeatureExtractor",
+  "image_processor_type": "ViTImageProcessor",
   "image_std": [
-    0.229,
-    0.224,
-    0.225
+    0.5,
+    0.5,
+    0.5
   ],
-  "resample": 3,
+  "resample": 2,
   "rescale_factor": 0.00392156862745098,
   "size": {
     "height": 224,

runs/Jan11_07-36-48_884473a9f125/events.out.tfevents.1704958690.884473a9f125.5987.13

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b659f3536afd6f2a79d3131e8d58ce14e048d87c46191c7a514e022f43564d6
+size 405

runs/Jan11_08-11-35_884473a9f125/events.out.tfevents.1704960712.884473a9f125.5987.14

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:677fe669792e570394ce0c7db69a10e3c40a073fde3c95e1f60a3af308b627ab
+size 4663

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 3.0,
-    "total_flos": 1.4093345754980352e+16,
-    "train_loss": 0.5933622784084744,
-    "train_runtime": 25.6802,
-    "train_samples_per_second": 22.079,
-    "train_steps_per_second": 0.35
+    "epoch": 5.0,
+    "total_flos": 2.348890959163392e+16,
+    "train_loss": 0.2989959716796875,
+    "train_runtime": 40.2987,
+    "train_samples_per_second": 23.45,
+    "train_steps_per_second": 0.372
 }

trainer_state.json

@@ -1,55 +1,79 @@
 {
-  "best_metric": 0.7727272727272727,
-  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat-kornia/checkpoint-9",
-  "epoch": 3.0,
+  "best_metric": 0.7272727272727273,
+  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat-kornia/checkpoint-6",
+  "epoch": 5.0,
   "eval_steps": 500,
-  "global_step": 9,
+  "global_step": 15,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.2727272727272727,
-      "eval_loss": 2.8753721714019775,
-      "eval_runtime": 0.2391,
-      "eval_samples_per_second": 91.995,
-      "eval_steps_per_second": 8.363,
+      "eval_accuracy": 0.5454545454545454,
+      "eval_loss": 1.666408658027649,
+      "eval_runtime": 0.2367,
+      "eval_samples_per_second": 92.955,
+      "eval_steps_per_second": 8.45,
       "step": 3
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.7272727272727273,
-      "eval_loss": 0.5872368812561035,
-      "eval_runtime": 0.2697,
-      "eval_samples_per_second": 81.585,
-      "eval_steps_per_second": 7.417,
+      "eval_loss": 0.7899118065834045,
+      "eval_runtime": 0.3107,
+      "eval_samples_per_second": 70.812,
+      "eval_steps_per_second": 6.437,
       "step": 6
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.7727272727272727,
-      "eval_loss": 0.5769956111907959,
-      "eval_runtime": 0.2361,
-      "eval_samples_per_second": 93.199,
-      "eval_steps_per_second": 8.473,
+      "eval_accuracy": 0.7272727272727273,
+      "eval_loss": 0.770358145236969,
+      "eval_runtime": 0.2398,
+      "eval_samples_per_second": 91.755,
+      "eval_steps_per_second": 8.341,
       "step": 9
     },
     {
-      "epoch": 3.0,
-      "step": 9,
-      "total_flos": 1.4093345754980352e+16,
-      "train_loss": 0.5933622784084744,
-      "train_runtime": 25.6802,
-      "train_samples_per_second": 22.079,
-      "train_steps_per_second": 0.35
+      "epoch": 3.33,
+      "learning_rate": 1.923076923076923e-05,
+      "loss": 0.3108,
+      "step": 10
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7272727272727273,
+      "eval_loss": 0.6269326210021973,
+      "eval_runtime": 0.2432,
+      "eval_samples_per_second": 90.452,
+      "eval_steps_per_second": 8.223,
+      "step": 12
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.6818181818181818,
+      "eval_loss": 0.644472062587738,
+      "eval_runtime": 0.2431,
+      "eval_samples_per_second": 90.505,
+      "eval_steps_per_second": 8.228,
+      "step": 15
+    },
+    {
+      "epoch": 5.0,
+      "step": 15,
+      "total_flos": 2.348890959163392e+16,
+      "train_loss": 0.2989959716796875,
+      "train_runtime": 40.2987,
+      "train_samples_per_second": 23.45,
+      "train_steps_per_second": 0.372
     }
   ],
   "logging_steps": 10,
-  "max_steps": 9,
-  "num_train_epochs": 3,
+  "max_steps": 15,
+  "num_train_epochs": 5,
   "save_steps": 500,
-  "total_flos": 1.4093345754980352e+16,
+  "total_flos": 2.348890959163392e+16,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:91266273ddd1721eb2e4f402a434c3d9fdf79c26e993dbc9f212a32c572151d8
+oid sha256:ac99b8d429f786afbe2368d73a7192ac7df9ec55a0e50f269048b862ba068f40
 size 4664