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: Segments
@@ -13,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Segments
 
-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 segments/sidewalk-semantic dataset.
 
 ## Model description
 

all_results.json

@@ -1,5 +1,5 @@
 {
-    "epoch": 0.99,
+    "epoch": 29.72,
     "eval_accuracy_construction-bridge": 0.0,
     "eval_accuracy_construction-building": 0.9044887536731845,
     "eval_accuracy_construction-door": 0.0,
@@ -77,8 +77,8 @@
     "eval_runtime": 166.5888,
     "eval_samples_per_second": 0.9,
     "eval_steps_per_second": 0.114,
-    "train_loss": 3.0431386119914503,
-    "train_runtime": 103.8023,
-    "train_samples_per_second": 8.189,
-    "train_steps_per_second": 0.511
+    "train_loss": 1.150932288019912,
+    "train_runtime": 3015.3294,
+    "train_samples_per_second": 8.457,
+    "train_steps_per_second": 0.527
 }

train_results.json

@@ -1,7 +1,7 @@
 {
-    "epoch": 0.99,
-    "train_loss": 3.0431386119914503,
-    "train_runtime": 103.8023,
-    "train_samples_per_second": 8.189,
-    "train_steps_per_second": 0.511
+    "epoch": 29.72,
+    "train_loss": 1.150932288019912,
+    "train_runtime": 3015.3294,
+    "train_samples_per_second": 8.457,
+    "train_steps_per_second": 0.527
 }

trainer_state.json

@@ -1,28 +1,46 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.9906542056074766,
+  "epoch": 29.7196261682243,
   "eval_steps": 500,
-  "global_step": 53,
+  "global_step": 1590,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.99,
-      "step": 53,
-      "total_flos": 1.4897641867444224e+16,
-      "train_loss": 3.0431386119914503,
-      "train_runtime": 103.8023,
-      "train_samples_per_second": 8.189,
-      "train_steps_per_second": 0.511
+      "epoch": 9.35,
+      "learning_rate": 3.4276729559748424e-05,
+      "loss": 1.6882,
+      "step": 500
+    },
+    {
+      "epoch": 18.69,
+      "learning_rate": 1.8553459119496856e-05,
+      "loss": 0.9858,
+      "step": 1000
+    },
+    {
+      "epoch": 28.04,
+      "learning_rate": 2.830188679245283e-06,
+      "loss": 0.8402,
+      "step": 1500
+    },
+    {
+      "epoch": 29.72,
+      "step": 1590,
+      "total_flos": 4.438724285646766e+17,
+      "train_loss": 1.150932288019912,
+      "train_runtime": 3015.3294,
+      "train_samples_per_second": 8.457,
+      "train_steps_per_second": 0.527
     }
   ],
   "logging_steps": 500,
-  "max_steps": 53,
-  "num_train_epochs": 1,
+  "max_steps": 1590,
+  "num_train_epochs": 30,
   "save_steps": 200,
-  "total_flos": 1.4897641867444224e+16,
+  "total_flos": 4.438724285646766e+17,
   "trial_name": null,
   "trial_params": null
 }