Upload MIMIC test evaluation results

README.md

@@ -34,7 +34,7 @@ metrics:
 - Project status: `Training in progress`
 - Release status: `Research preview checkpoint`
 - Current checkpoint status: `Not final`
-- Training completion toward planned run: `98.26%` (`2.948` / `3` epochs)
+- Training completion toward planned run: `100.00%` (`4.000` / `3` epochs)
 - Current published metrics are intermediate and will change as training continues.
 
 ## Overview
@@ -79,12 +79,12 @@ For local inference instructions, go to the [Inference](#inference) section.
 - Scheduler: `cosine`
 - Warmup steps: `5114`
 - Weight decay: `0.01`
-- Steps completed: `100480`
+- Steps completed: `102264`
 - Planned total steps: `102276`
-- Images seen: `803922`
-- Total training time: `23.1894` hours
+- Images seen: `818196`
+- Total training time: `23.5798` hours
 - Hardware: `NVIDIA GeForce RTX 5070`
-- Final train loss: `2.1738`
+- Final train loss: `1.1683`
 - Validation loss: `1.3692`
 
 ## MIMIC Test Results
@@ -96,13 +96,13 @@ Frontal-only evaluation using `PA/AP` studies only.
 | Metric | Value |
 | --- | --- |
 | Number of studies | `3041` |
-| RadGraph F1 | `0.0921` |
-| RadGraph entity F1 | `0.1404` |
-| RadGraph relation F1 | `0.1250` |
-| CheXpert F1 14-micro | `0.1840` |
-| CheXpert F1 5-micro | `0.2200` |
-| CheXpert F1 14-macro | `0.1101` |
-| CheXpert F1 5-macro | `0.1648` |
+| RadGraph F1 | `0.0918` |
+| RadGraph entity F1 | `0.1399` |
+| RadGraph relation F1 | `0.1246` |
+| CheXpert F1 14-micro | `0.1829` |
+| CheXpert F1 5-micro | `0.2183` |
+| CheXpert F1 14-macro | `0.1095` |
+| CheXpert F1 5-macro | `0.1634` |
 
 ### Final Completed Training Results
 
@@ -182,13 +182,13 @@ print(report)
 - Dataset: `MIMIC-CXR test`
 - View filter: `frontal-only (PA/AP)`
 - Number of examples: `3041`
-- CheXpert F1 14-micro: `0.1840`
-- CheXpert F1 5-micro: `0.2200`
-- CheXpert F1 14-macro: `0.1101`
-- CheXpert F1 5-macro: `0.1648`
-- RadGraph F1: `0.0921`
-- RadGraph entity F1: `0.1404`
-- RadGraph relation F1: `0.1250`
+- CheXpert F1 14-micro: `0.1829`
+- CheXpert F1 5-micro: `0.2183`
+- CheXpert F1 14-macro: `0.1095`
+- CheXpert F1 5-macro: `0.1634`
+- RadGraph F1: `0.0918`
+- RadGraph entity F1: `0.1399`
+- RadGraph relation F1: `0.1246`
 - RadGraph available: `True`
 - RadGraph error: `None`
 

evaluations/mimic_test_metrics.json

@@ -3,31 +3,31 @@
   "dataset": "mimic-cxr",
   "view_filter": "frontal-only (PA/AP)",
   "num_examples": 3041,
-  "chexpert_f1_14_micro": 0.18400222191362306,
-  "chexpert_f1_5_micro": 0.21997816593886463,
-  "chexpert_f1_14_macro": 0.11010850567969291,
-  "chexpert_f1_5_macro": 0.16475190455586672,
-  "chexpert_f1_micro": 0.18400222191362306,
-  "chexpert_f1_macro": 0.11010850567969291,
+  "chexpert_f1_14_micro": 0.18291666666666664,
+  "chexpert_f1_5_micro": 0.21831082003001773,
+  "chexpert_f1_14_macro": 0.10945797832551928,
+  "chexpert_f1_5_macro": 0.1633553219570594,
+  "chexpert_f1_micro": 0.18291666666666664,
+  "chexpert_f1_macro": 0.10945797832551928,
   "chexpert_per_label_f1": {
     "Enlarged Cardiomediastinum": 0.0,
-    "Cardiomegaly": 0.0998914223669924,
+    "Cardiomegaly": 0.10195227765726682,
     "Lung Opacity": 0.0020470829068577278,
     "Lung Lesion": 0.0,
-    "Edema": 0.28379287155346333,
-    "Consolidation": 0.06440677966101695,
-    "Pneumonia": 0.14285714285714285,
-    "Atelectasis": 0.045454545454545456,
-    "Pneumothorax": 0.031572556162720096,
-    "Pleural Effusion": 0.3302139037433155,
+    "Edema": 0.2789757412398922,
+    "Consolidation": 0.06424344885883347,
+    "Pneumonia": 0.14311926605504585,
+    "Atelectasis": 0.0428380187416332,
+    "Pneumothorax": 0.030358227079538558,
+    "Pleural Effusion": 0.32876712328767127,
     "Pleural Other": 0.0,
-    "Fracture": 0.06331877729257641,
-    "Support Devices": 0.4779639975170701,
+    "Fracture": 0.0633879781420765,
+    "Support Devices": 0.4767225325884544,
     "No Finding": 0.0
   },
-  "radgraph_f1": 0.09213455645553452,
-  "radgraph_f1_entity": 0.14037285346616235,
-  "radgraph_f1_relation": 0.12500236867970504,
+  "radgraph_f1": 0.09181957971495504,
+  "radgraph_f1_entity": 0.13993790644379023,
+  "radgraph_f1_relation": 0.12464719867951028,
   "radgraph_available": true,
   "radgraph_error": null
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:406e459b379055f5d9796be4a9d2ee68641d7947388384cd2405a7aac138a836
+oid sha256:3ad12d9319ce0de4ae32d6b954079fee71ed77c3e4f94e3a7e17db14f98f6fe2
 size 1152540320

run_summary.json

@@ -1,18 +1,18 @@
 {
   "method": "lora_adamw",
   "run_name": "full_3_epoch_mask_run",
-  "steps": 100480,
-  "epochs_completed": 2,
-  "epoch_index": 2,
+  "steps": 102264,
+  "epochs_completed": 3,
+  "epoch_index": 3,
   "target_epochs": 3,
-  "progress_epochs": 2.9476629071762757,
-  "training_completion_percent": 98.25543023920919,
-  "elapsed_seconds": 83481.91898369994,
-  "images_seen": 803922,
-  "train_loss_last": 2.173830270767212,
-  "train_loss_mean": 1.8575981099994647,
-  "val_loss": 1.369218921661377,
-  "images_per_second": 9.629893631900913,
+  "progress_epochs": 4.0,
+  "training_completion_percent": 100.0,
+  "elapsed_seconds": 84887.16960079994,
+  "images_seen": 818196,
+  "train_loss_last": 1.1683106422424316,
+  "train_loss_mean": 1.8536666353336062,
+  "val_loss": 1.3691736698150634,
+  "images_per_second": 9.638629769937454,
   "trainable_params": 1106688,
   "vision_model_name": "facebook/dinov3-vits16-pretrain-lvd1689m",
   "text_model_name": "gpt2",
@@ -33,9 +33,9 @@
   "seed": 42,
   "resume_supported": true,
   "checkpoint_every_n_steps": 1000,
-  "cumulative_loss_sum": 1493363.9877869897,
-  "cumulative_loss_count": 803922,
-  "completed": false,
+  "cumulative_loss_sum": 1516662.6263634153,
+  "cumulative_loss_count": 818196,
+  "completed": true,
   "target_duration_seconds": 3600,
   "target_duration_mode": "per_invocation",
   "train_datasets": "CheXpert, MIMIC-CXR",
@@ -45,31 +45,31 @@
     "dataset": "mimic-cxr",
     "view_filter": "frontal-only (PA/AP)",
     "num_examples": 3041,
-    "chexpert_f1_14_micro": 0.18400222191362306,
-    "chexpert_f1_5_micro": 0.21997816593886463,
-    "chexpert_f1_14_macro": 0.11010850567969291,
-    "chexpert_f1_5_macro": 0.16475190455586672,
-    "chexpert_f1_micro": 0.18400222191362306,
-    "chexpert_f1_macro": 0.11010850567969291,
+    "chexpert_f1_14_micro": 0.18291666666666664,
+    "chexpert_f1_5_micro": 0.21831082003001773,
+    "chexpert_f1_14_macro": 0.10945797832551928,
+    "chexpert_f1_5_macro": 0.1633553219570594,
+    "chexpert_f1_micro": 0.18291666666666664,
+    "chexpert_f1_macro": 0.10945797832551928,
     "chexpert_per_label_f1": {
       "Enlarged Cardiomediastinum": 0.0,
-      "Cardiomegaly": 0.0998914223669924,
+      "Cardiomegaly": 0.10195227765726682,
       "Lung Opacity": 0.0020470829068577278,
       "Lung Lesion": 0.0,
-      "Edema": 0.28379287155346333,
-      "Consolidation": 0.06440677966101695,
-      "Pneumonia": 0.14285714285714285,
-      "Atelectasis": 0.045454545454545456,
-      "Pneumothorax": 0.031572556162720096,
-      "Pleural Effusion": 0.3302139037433155,
+      "Edema": 0.2789757412398922,
+      "Consolidation": 0.06424344885883347,
+      "Pneumonia": 0.14311926605504585,
+      "Atelectasis": 0.0428380187416332,
+      "Pneumothorax": 0.030358227079538558,
+      "Pleural Effusion": 0.32876712328767127,
       "Pleural Other": 0.0,
-      "Fracture": 0.06331877729257641,
-      "Support Devices": 0.4779639975170701,
+      "Fracture": 0.0633879781420765,
+      "Support Devices": 0.4767225325884544,
       "No Finding": 0.0
     },
-    "radgraph_f1": 0.09213455645553452,
-    "radgraph_f1_entity": 0.14037285346616235,
-    "radgraph_f1_relation": 0.12500236867970504,
+    "radgraph_f1": 0.09181957971495504,
+    "radgraph_f1_entity": 0.13993790644379023,
+    "radgraph_f1_relation": 0.12464719867951028,
     "radgraph_available": true,
     "radgraph_error": null
   }