End of training

README.md

@@ -14,6 +14,8 @@ should probably proofread and complete it, then remove this comment. -->
 # clip_finetune_0207
 
 This model is a fine-tuned version of [openai/clip-vit-base-patch32](https://huggingface.co/openai/clip-vit-base-patch32) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3562
 
 ## Model description
 

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 10.0,
-    "eval_loss": 0.34109967947006226,
-    "eval_runtime": 382.2208,
-    "eval_samples_per_second": 15.219,
-    "eval_steps_per_second": 1.905,
+    "eval_loss": 0.3561580181121826,
+    "eval_runtime": 381.0514,
+    "eval_samples_per_second": 15.266,
+    "eval_steps_per_second": 1.911,
     "total_flos": 3.031314932000124e+16,
-    "train_loss": 0.4382253171182146,
-    "train_runtime": 35381.4579,
-    "train_samples_per_second": 14.729,
-    "train_steps_per_second": 0.058
+    "train_loss": 0.4388976242028031,
+    "train_runtime": 38108.8591,
+    "train_samples_per_second": 13.675,
+    "train_steps_per_second": 0.054
 }

eval_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 10.0,
-    "eval_loss": 0.34109967947006226,
-    "eval_runtime": 382.2208,
-    "eval_samples_per_second": 15.219,
-    "eval_steps_per_second": 1.905
+    "eval_loss": 0.3561580181121826,
+    "eval_runtime": 381.0514,
+    "eval_samples_per_second": 15.266,
+    "eval_steps_per_second": 1.911
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
     "total_flos": 3.031314932000124e+16,
-    "train_loss": 0.4382253171182146,
-    "train_runtime": 35381.4579,
-    "train_samples_per_second": 14.729,
-    "train_steps_per_second": 0.058
+    "train_loss": 0.4388976242028031,
+    "train_runtime": 38108.8591,
+    "train_samples_per_second": 13.675,
+    "train_steps_per_second": 0.054
 }

trainer_state.json

@@ -17,33 +17,33 @@
     },
     {
       "epoch": 4.901960784313726,
-      "grad_norm": 32.49726867675781,
+      "grad_norm": 32.042449951171875,
       "learning_rate": 3e-05,
-      "loss": 0.4341,
+      "loss": 0.4359,
       "step": 1000
     },
     {
       "epoch": 7.352941176470588,
-      "grad_norm": 9.985998153686523,
+      "grad_norm": 16.227807998657227,
       "learning_rate": 1.557692307692308e-05,
-      "loss": 0.2037,
+      "loss": 0.2055,
       "step": 1500
     },
     {
       "epoch": 9.803921568627452,
-      "grad_norm": 7.020397663116455,
+      "grad_norm": 7.223657131195068,
       "learning_rate": 1.153846153846154e-06,
-      "loss": 0.0495,
+      "loss": 0.0488,
       "step": 2000
     },
     {
       "epoch": 10.0,
       "step": 2040,
       "total_flos": 3.031314932000124e+16,
-      "train_loss": 0.4382253171182146,
-      "train_runtime": 35381.4579,
-      "train_samples_per_second": 14.729,
-      "train_steps_per_second": 0.058
+      "train_loss": 0.4388976242028031,
+      "train_runtime": 38108.8591,
+      "train_samples_per_second": 13.675,
+      "train_steps_per_second": 0.054
     }
   ],
   "logging_steps": 500,