Update README.md

README.md

@@ -29,7 +29,7 @@ A small convolutional neural network, written and trained from scratch in PyTorc
 ## Data
 
 - 297 spectrogram images: 101 background, 99 helicopter, 97 drone.
-- Spectrograms were generated from [source audio: where the recordings came from, and the tool or settings used to convert them].
+- Spectrograms were generated from dataset provided by the London Defence Hackathon community on GitHub
 - Random 60/20/20 train/validation/test split (`random_state=42`), about 178 / 59 / 60 images.
 
 ## Training
@@ -40,13 +40,13 @@ A small convolutional neural network, written and trained from scratch in PyTorc
 
 ## Results (60 held-out test spectrograms)
 
-- Test accuracy: [xx.x]%
-- Weighted F1: [0.xx] (precision [0.xx], recall [0.xx])
+- Test accuracy: 86.7%
+- Weighted F1: 0.86 (precision 0.89, recall 0.86%)
 - Per-class one-vs-rest ROC-AUC: helicopter 1.00, drone 0.97, background 0.96
 
 ## Limitations
 
-- Small dataset, and a small test set (60 images), so these figures are indicative rather than precise.
+- Small dataset of 297 recordings and a small test set (60 images) so these figures are indicative rather than precise.
 - Training loss is very low while validation accuracy is lower, which suggests overfitting. No data augmentation or regularisation was used.
 - The split is random at image level. [State whether all spectrograms come from separate recordings. If some share a source recording, say so: near-duplicate clips in train and test would inflate the scores, and splitting by recording is the planned next step.]
 - Not tested on recordings from new environments, microphones or noise conditions.