VulcanRaven
/

ChartQA-smolvlm

Image-Text-to-Text

vision-language

document-understanding

Model card Files Files and versions

VulcanRaven commited on Mar 15

Commit

f0080c8

·

verified ·

1 Parent(s): ca66743

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -112,7 +112,7 @@ processor = AutoProcessor.from_pretrained("VulcanRaven/ChartQA-smolvlm")
 ## Training details
-- **Hardware**: NVIDIA Tesla T4 (16 GB) — resource-constrained environment
 - **Data split**: 80% train / 20% validation (from original train), full original test set
 - **Validation**: 50-batch subset evaluated after each epoch for speed
 - **Best checkpoint**: Saved based on highest Relaxed Accuracy on validation set

 ## Training details
+- **Hardware**: NVIDIA Tesla T4
 - **Data split**: 80% train / 20% validation (from original train), full original test set
 - **Validation**: 50-batch subset evaluated after each epoch for speed
 - **Best checkpoint**: Saved based on highest Relaxed Accuracy on validation set