End of training

README.md

@@ -2,6 +2,8 @@
 license: other
 base_model: nvidia/mit-b0
 tags:
+- image-segmentation
+- vision
 - generated_from_trainer
 model-index:
 - name: segformer-finetuned-obb-1k-steps
@@ -13,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # segformer-finetuned-obb-1k-steps
 
-This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on an unknown dataset.
+This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on the Spatiallysaying/obb dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.0511
 - Mean Iou: 0.2238

all_results.json

@@ -0,0 +1,19 @@
+{
+    "epoch": 5.780346820809249,
+    "eval_accuracy__backgound_": NaN,
+    "eval_accuracy_rwy_obb": 0.4476643255897713,
+    "eval_iou__backgound_": 0.0,
+    "eval_iou_rwy_obb": 0.4476643255897713,
+    "eval_loss": 0.051128923892974854,
+    "eval_mean_accuracy": 0.4476643255897713,
+    "eval_mean_iou": 0.22383216279488566,
+    "eval_overall_accuracy": 0.4476643255897713,
+    "eval_runtime": 67.0374,
+    "eval_samples_per_second": 1.298,
+    "eval_steps_per_second": 0.656,
+    "total_flos": 3.5055911043072e+16,
+    "train_loss": 0.10499012756347656,
+    "train_runtime": 1942.3376,
+    "train_samples_per_second": 1.03,
+    "train_steps_per_second": 0.515
+}

eval_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 5.780346820809249,
+    "eval_accuracy__backgound_": NaN,
+    "eval_accuracy_rwy_obb": 0.4476643255897713,
+    "eval_iou__backgound_": 0.0,
+    "eval_iou_rwy_obb": 0.4476643255897713,
+    "eval_loss": 0.051128923892974854,
+    "eval_mean_accuracy": 0.4476643255897713,
+    "eval_mean_iou": 0.22383216279488566,
+    "eval_overall_accuracy": 0.4476643255897713,
+    "eval_runtime": 67.0374,
+    "eval_samples_per_second": 1.298,
+    "eval_steps_per_second": 0.656
+}

runs/Jul14_01-00-14_bc7841011100/events.out.tfevents.1720920842.bc7841011100.2840.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfb30396a82d52f79ba6fdff4f06f6e291e59019847e10d656c979203f11b2fb
+size 766

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 5.780346820809249,
+    "total_flos": 3.5055911043072e+16,
+    "train_loss": 0.10499012756347656,
+    "train_runtime": 1942.3376,
+    "train_samples_per_second": 1.03,
+    "train_steps_per_second": 0.515
+}

trainer_state.json

@@ -0,0 +1,202 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.780346820809249,
+  "eval_steps": 500,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5780346820809249,
+      "grad_norm": 1.5270891189575195,
+      "learning_rate": 5.401e-05,
+      "loss": 0.3927,
+      "step": 100
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy__backgound_": NaN,
+      "eval_accuracy_rwy_obb": 0.3179992796686476,
+      "eval_iou__backgound_": 0.0,
+      "eval_iou_rwy_obb": 0.3179992796686476,
+      "eval_loss": 0.10957548767328262,
+      "eval_mean_accuracy": 0.3179992796686476,
+      "eval_mean_iou": 0.1589996398343238,
+      "eval_overall_accuracy": 0.3179992796686476,
+      "eval_runtime": 64.8551,
+      "eval_samples_per_second": 1.341,
+      "eval_steps_per_second": 0.678,
+      "step": 173
+    },
+    {
+      "epoch": 1.1560693641618498,
+      "grad_norm": 0.8130283355712891,
+      "learning_rate": 4.8020000000000004e-05,
+      "loss": 0.1569,
+      "step": 200
+    },
+    {
+      "epoch": 1.7341040462427746,
+      "grad_norm": 1.5858874320983887,
+      "learning_rate": 4.203e-05,
+      "loss": 0.0969,
+      "step": 300
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy__backgound_": NaN,
+      "eval_accuracy_rwy_obb": 0.4224131100306141,
+      "eval_iou__backgound_": 0.0,
+      "eval_iou_rwy_obb": 0.4224131100306141,
+      "eval_loss": 0.0703619122505188,
+      "eval_mean_accuracy": 0.4224131100306141,
+      "eval_mean_iou": 0.21120655501530705,
+      "eval_overall_accuracy": 0.4224131100306141,
+      "eval_runtime": 66.7554,
+      "eval_samples_per_second": 1.303,
+      "eval_steps_per_second": 0.659,
+      "step": 346
+    },
+    {
+      "epoch": 2.3121387283236996,
+      "grad_norm": 0.44887179136276245,
+      "learning_rate": 3.604e-05,
+      "loss": 0.076,
+      "step": 400
+    },
+    {
+      "epoch": 2.8901734104046244,
+      "grad_norm": 0.6084420084953308,
+      "learning_rate": 3.0050000000000002e-05,
+      "loss": 0.0651,
+      "step": 500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy__backgound_": NaN,
+      "eval_accuracy_rwy_obb": 0.43714928867278946,
+      "eval_iou__backgound_": 0.0,
+      "eval_iou_rwy_obb": 0.43714928867278946,
+      "eval_loss": 0.059844404458999634,
+      "eval_mean_accuracy": 0.43714928867278946,
+      "eval_mean_iou": 0.21857464433639473,
+      "eval_overall_accuracy": 0.43714928867278946,
+      "eval_runtime": 69.5674,
+      "eval_samples_per_second": 1.251,
+      "eval_steps_per_second": 0.632,
+      "step": 519
+    },
+    {
+      "epoch": 3.468208092485549,
+      "grad_norm": 0.5608803629875183,
+      "learning_rate": 2.406e-05,
+      "loss": 0.0576,
+      "step": 600
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy__backgound_": NaN,
+      "eval_accuracy_rwy_obb": 0.44997839005942736,
+      "eval_iou__backgound_": 0.0,
+      "eval_iou_rwy_obb": 0.44997839005942736,
+      "eval_loss": 0.053011514246463776,
+      "eval_mean_accuracy": 0.44997839005942736,
+      "eval_mean_iou": 0.22498919502971368,
+      "eval_overall_accuracy": 0.44997839005942736,
+      "eval_runtime": 71.7418,
+      "eval_samples_per_second": 1.213,
+      "eval_steps_per_second": 0.613,
+      "step": 692
+    },
+    {
+      "epoch": 4.046242774566474,
+      "grad_norm": 1.257560133934021,
+      "learning_rate": 1.807e-05,
+      "loss": 0.0569,
+      "step": 700
+    },
+    {
+      "epoch": 4.624277456647399,
+      "grad_norm": 0.7843059301376343,
+      "learning_rate": 1.2079999999999998e-05,
+      "loss": 0.0531,
+      "step": 800
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy__backgound_": NaN,
+      "eval_accuracy_rwy_obb": 0.4424491265982352,
+      "eval_iou__backgound_": 0.0,
+      "eval_iou_rwy_obb": 0.4424491265982352,
+      "eval_loss": 0.05286609008908272,
+      "eval_mean_accuracy": 0.4424491265982352,
+      "eval_mean_iou": 0.2212245632991176,
+      "eval_overall_accuracy": 0.4424491265982352,
+      "eval_runtime": 71.2619,
+      "eval_samples_per_second": 1.221,
+      "eval_steps_per_second": 0.617,
+      "step": 865
+    },
+    {
+      "epoch": 5.202312138728324,
+      "grad_norm": 1.10871422290802,
+      "learning_rate": 6.0899999999999984e-06,
+      "loss": 0.048,
+      "step": 900
+    },
+    {
+      "epoch": 5.780346820809249,
+      "grad_norm": 0.6045771837234497,
+      "learning_rate": 1e-07,
+      "loss": 0.0467,
+      "step": 1000
+    },
+    {
+      "epoch": 5.780346820809249,
+      "eval_accuracy__backgound_": NaN,
+      "eval_accuracy_rwy_obb": 0.4476643255897713,
+      "eval_iou__backgound_": 0.0,
+      "eval_iou_rwy_obb": 0.4476643255897713,
+      "eval_loss": 0.051128923892974854,
+      "eval_mean_accuracy": 0.4476643255897713,
+      "eval_mean_iou": 0.22383216279488566,
+      "eval_overall_accuracy": 0.4476643255897713,
+      "eval_runtime": 68.1508,
+      "eval_samples_per_second": 1.277,
+      "eval_steps_per_second": 0.646,
+      "step": 1000
+    },
+    {
+      "epoch": 5.780346820809249,
+      "step": 1000,
+      "total_flos": 3.5055911043072e+16,
+      "train_loss": 0.10499012756347656,
+      "train_runtime": 1942.3376,
+      "train_samples_per_second": 1.03,
+      "train_steps_per_second": 0.515
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 1000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.5055911043072e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}