Spaces:
Running
Running
“Namhyun-Kim”
commited on
Commit
·
d0f63ee
1
Parent(s):
ebad958
Add t-SNE usage note for modulation-by-SNR sweeps
Browse files
README.md
CHANGED
|
@@ -30,6 +30,6 @@ Interactive lab for exploring **LWM spectrogram embeddings** on Sub-6 GHz I/Q ba
|
|
| 30 |
|
| 31 |
## Tab Cheat Sheet
|
| 32 |
- **Spectrograms:** Inspect raw 128×128 Sub-6 GHz I/Q baseband spectrograms per technology/SNR/modulation/mobility before feature extraction.
|
| 33 |
-
- **t-SNE Analysis:** Recreate the SNR-ordered scatter plots from `plot/plot_tsne.py` with balanced sampling across SNRs.
|
| 34 |
- **Modulation Classification:** Benchmark a lightweight k-NN probe on LWM embeddings vs. raw inputs for each technology.
|
| 35 |
- **Joint SNR/Doppler Evaluation:** Compare cached MoE embeddings and raw spectrograms on the 14-way SNR/mobility task.
|
|
|
|
| 30 |
|
| 31 |
## Tab Cheat Sheet
|
| 32 |
- **Spectrograms:** Inspect raw 128×128 Sub-6 GHz I/Q baseband spectrograms per technology/SNR/modulation/mobility before feature extraction.
|
| 33 |
+
- **t-SNE Analysis:** Recreate the SNR-ordered scatter plots from `plot/plot_tsne.py` with balanced sampling across SNRs. Example: hold SNR fixed and color by modulation to see how modulation clusters separate (or collapse) as SNR changes.
|
| 34 |
- **Modulation Classification:** Benchmark a lightweight k-NN probe on LWM embeddings vs. raw inputs for each technology.
|
| 35 |
- **Joint SNR/Doppler Evaluation:** Compare cached MoE embeddings and raw spectrograms on the 14-way SNR/mobility task.
|