allen-ajith commited on
Commit
f990e26
·
verified ·
1 Parent(s): ccf32dd

Add training summary

Browse files
Files changed (1) hide show
  1. README.md +31 -0
README.md ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # resnet50
2
+
3
+ **Best Validation Accuracy:** `0.8733`
4
+
5
+ ## Metadata
6
+ - **Model Name**: `resnet50`
7
+ - **Optimizer**: `adamw`
8
+ - **Scheduler**: `cosine`
9
+ - **Weight Decay**: `0.05`
10
+ - **Warmup Epochs**: `3`
11
+ - **Patience**: `5`
12
+ - **Amp**: `True`
13
+ - **Seed**: `42`
14
+ - **Batch Size**: `32`
15
+ - **Initial Lr**: `7.5e-05`
16
+ - **Total Epochs Ran**: `1`
17
+ - **Early Stopped**: `False`
18
+ - **Training Time Seconds**: `2375.9705023765564`
19
+ - **Num Parameters**: `22067009`
20
+ - **Device**: `NVIDIA A100-SXM4-40GB`
21
+ - **Run Id**: `resnet-test`
22
+
23
+ ## Training Configuration
24
+ - Epochs: `1`
25
+ - Batch size: `32`
26
+ - Learning rate (initial): `7.5e-05`
27
+
28
+ ## Training Logs (Per Epoch)
29
+ | Epoch | Train Loss | Train Acc | Val Loss | Val Acc | LR |
30
+ |-------|------------|-----------|----------|---------|----|
31
+ | 1 | 0.2911 | 0.8787 | 0.3048 | 0.8733 | 0.000025 |