🍻 cheers

README.md

@@ -1,7 +1,11 @@
 ---
+language:
+- eng
 license: apache-2.0
-base_model: lombardata/dinov2-large-2024_01_24-with_data_aug_batch-size32_epochs85_freeze
+base_model: facebook/dinov2-large
 tags:
+- multilabel-image-classification
+- multilabel
 - generated_from_trainer
 metrics:
 - accuracy
@@ -15,13 +19,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 # dinov2-large-2024_01_24-with_data_aug_batch-size32_epochs85_freeze
 
-This model is a fine-tuned version of [lombardata/dinov2-large-2024_01_24-with_data_aug_batch-size32_epochs85_freeze](https://huggingface.co/lombardata/dinov2-large-2024_01_24-with_data_aug_batch-size32_epochs85_freeze) on the None dataset.
+This model is a fine-tuned version of [facebook/dinov2-large](https://huggingface.co/facebook/dinov2-large) on the multilabel_complete_dataset dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0871
-- F1 Micro: 0.8657
-- F1 Macro: 0.8277
-- Roc Auc: 0.9141
-- Accuracy: 0.5807
+- Loss: 0.0864
+- F1 Micro: 0.8668
+- F1 Macro: 0.8381
+- Roc Auc: 0.9138
+- Accuracy: 0.5805
 - Learning Rate: 0.0000
 
 ## Model description

all_results.json

@@ -5,12 +5,13 @@
     "eval_f1_micro": 0.866814335207588,
     "eval_loss": 0.0863719955086708,
     "eval_roc_auc": 0.9138424851830095,
-    "eval_runtime": 679.2028,
-    "eval_samples_per_second": 4.198,
-    "eval_steps_per_second": 0.133,
+    "eval_runtime": 679.6067,
+    "eval_samples_per_second": 4.195,
+    "eval_steps_per_second": 0.132,
     "learning_rate": 1.0000000000000002e-07,
-    "train_loss": 0.10122811819873383,
-    "train_runtime": 238033.4867,
-    "train_samples_per_second": 3.13,
-    "train_steps_per_second": 0.098
+    "total_flos": 1.1045912459199104e+21,
+    "train_loss": 0.0019856175725272715,
+    "train_runtime": 5622.9029,
+    "train_samples_per_second": 132.483,
+    "train_steps_per_second": 4.142
 }

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "lombardata/dinov2-large-2024_01_24-with_data_aug_batch-size32_epochs85_freeze",
+  "_name_or_path": "facebook/dinov2-large2024_01_28",
   "apply_layernorm": true,
   "architectures": [
     "NewheadDinov2ForImageClassification"

eval_results.json

@@ -5,8 +5,8 @@
     "eval_f1_micro": 0.866814335207588,
     "eval_loss": 0.0863719955086708,
     "eval_roc_auc": 0.9138424851830095,
-    "eval_runtime": 679.2028,
-    "eval_samples_per_second": 4.198,
-    "eval_steps_per_second": 0.133,
+    "eval_runtime": 679.6067,
+    "eval_samples_per_second": 4.195,
+    "eval_steps_per_second": 0.132,
     "learning_rate": 1.0000000000000002e-07
 }

runs/Jan28_10-47-53_datavisu2/events.out.tfevents.1706441601.datavisu2.7474.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e875eb6c80a000579f1018c75b0b70c031cf79ce8c857c410ecabeab7f620eb
+size 634

train_results.json

@@ -1,8 +1,9 @@
 {
     "epoch": 85.0,
     "learning_rate": 1.0000000000000002e-07,
-    "train_loss": 0.10122811819873383,
-    "train_runtime": 238033.4867,
-    "train_samples_per_second": 3.13,
-    "train_steps_per_second": 0.098
+    "total_flos": 1.1045912459199104e+21,
+    "train_loss": 0.0019856175725272715,
+    "train_runtime": 5622.9029,
+    "train_samples_per_second": 132.483,
+    "train_steps_per_second": 4.142
 }

trainer_state.json

@@ -1360,7 +1360,7 @@
     {
       "epoch": 83.94,
       "learning_rate": 1.0000000000000002e-07,
-      "loss": 0.0842,
+      "loss": 0.0837,
       "step": 23000
     },
     {
@@ -1370,9 +1370,9 @@
       "eval_f1_micro": 0.8627316009866345,
       "eval_loss": 0.08810650557279587,
       "eval_roc_auc": 0.9079679208453217,
-      "eval_runtime": 681.9462,
-      "eval_samples_per_second": 4.214,
-      "eval_steps_per_second": 0.132,
+      "eval_runtime": 687.4756,
+      "eval_samples_per_second": 4.181,
+      "eval_steps_per_second": 0.131,
       "learning_rate": 1.0000000000000002e-07,
       "step": 23016
     },
@@ -1383,8 +1383,8 @@
       "eval_f1_micro": 0.8657459814353634,
       "eval_loss": 0.08707784116268158,
       "eval_roc_auc": 0.9141406112899818,
-      "eval_runtime": 686.2064,
-      "eval_samples_per_second": 4.188,
+      "eval_runtime": 688.8064,
+      "eval_samples_per_second": 4.172,
       "eval_steps_per_second": 0.131,
       "learning_rate": 1.0000000000000002e-07,
       "step": 23290
@@ -1394,10 +1394,10 @@
       "learning_rate": 1.0000000000000002e-07,
       "step": 23290,
       "total_flos": 1.1045912459199104e+21,
-      "train_loss": 0.10122811819873383,
-      "train_runtime": 238033.4867,
-      "train_samples_per_second": 3.13,
-      "train_steps_per_second": 0.098
+      "train_loss": 0.0019856175725272715,
+      "train_runtime": 5622.9029,
+      "train_samples_per_second": 132.483,
+      "train_steps_per_second": 4.142
     }
   ],
   "logging_steps": 500,