Deploy eval explorer dashboard
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- .gitattributes +4 -0
- README.md +15 -4
- index.html +0 -0
- per_query_shards/NanoArguAna.json +0 -0
- per_query_shards/NanoClimateFEVER.json +0 -0
- per_query_shards/NanoDBPedia.json +3 -0
- per_query_shards/NanoFEVER.json +0 -0
- per_query_shards/NanoFiQA2018.json +0 -0
- per_query_shards/NanoHotpotQA.json +0 -0
- per_query_shards/NanoMSMARCO.json +0 -0
- per_query_shards/NanoNFCorpus.json +3 -0
- per_query_shards/NanoNQ.json +0 -0
- per_query_shards/NanoQuoraRetrieval.json +0 -0
- per_query_shards/NanoSCIDOCS.json +3 -0
- per_query_shards/NanoSciFact.json +0 -0
- per_query_shards/NanoTouche2020.json +3 -0
- plotly.min.js +0 -0
- significance_shards/NanoArguAna__accuracy-1.json +0 -0
- significance_shards/NanoArguAna__accuracy-10.json +0 -0
- significance_shards/NanoArguAna__accuracy-3.json +0 -0
- significance_shards/NanoArguAna__accuracy-5.json +0 -0
- significance_shards/NanoArguAna__mrr-1.json +0 -0
- significance_shards/NanoArguAna__mrr-10.json +0 -0
- significance_shards/NanoArguAna__mrr-3.json +0 -0
- significance_shards/NanoArguAna__mrr-5.json +0 -0
- significance_shards/NanoArguAna__ndcg-1.json +0 -0
- significance_shards/NanoArguAna__ndcg-10.json +0 -0
- significance_shards/NanoArguAna__ndcg-3.json +0 -0
- significance_shards/NanoArguAna__ndcg-5.json +0 -0
- significance_shards/NanoArguAna__recall-1.json +0 -0
- significance_shards/NanoArguAna__recall-10.json +0 -0
- significance_shards/NanoArguAna__recall-3.json +0 -0
- significance_shards/NanoArguAna__recall-5.json +0 -0
- significance_shards/NanoClimateFEVER__accuracy-1.json +0 -0
- significance_shards/NanoClimateFEVER__accuracy-10.json +0 -0
- significance_shards/NanoClimateFEVER__accuracy-3.json +0 -0
- significance_shards/NanoClimateFEVER__accuracy-5.json +0 -0
- significance_shards/NanoClimateFEVER__mrr-1.json +0 -0
- significance_shards/NanoClimateFEVER__mrr-10.json +0 -0
- significance_shards/NanoClimateFEVER__mrr-3.json +0 -0
- significance_shards/NanoClimateFEVER__mrr-5.json +0 -0
- significance_shards/NanoClimateFEVER__ndcg-1.json +0 -0
- significance_shards/NanoClimateFEVER__ndcg-10.json +0 -0
- significance_shards/NanoClimateFEVER__ndcg-3.json +0 -0
- significance_shards/NanoClimateFEVER__ndcg-5.json +0 -0
- significance_shards/NanoClimateFEVER__recall-1.json +0 -0
- significance_shards/NanoClimateFEVER__recall-10.json +0 -0
- significance_shards/NanoClimateFEVER__recall-3.json +0 -0
- significance_shards/NanoClimateFEVER__recall-5.json +0 -0
- significance_shards/NanoDBPedia__accuracy-1.json +0 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
per_query_shards/NanoDBPedia.json filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
per_query_shards/NanoNFCorpus.json filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
per_query_shards/NanoSCIDOCS.json filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
per_query_shards/NanoTouche2020.json filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
|
@@ -1,10 +1,21 @@
|
|
| 1 |
---
|
| 2 |
title: Embedding Compression Eval
|
| 3 |
-
emoji:
|
| 4 |
-
colorFrom:
|
| 5 |
-
colorTo:
|
| 6 |
sdk: static
|
|
|
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
title: Embedding Compression Eval
|
| 3 |
+
emoji: 📊
|
| 4 |
+
colorFrom: indigo
|
| 5 |
+
colorTo: purple
|
| 6 |
sdk: static
|
| 7 |
+
app_file: index.html
|
| 8 |
pinned: false
|
| 9 |
---
|
| 10 |
|
| 11 |
+
# Embedding Compression Eval Explorer
|
| 12 |
+
|
| 13 |
+
Interactive NanoBEIR evaluation dashboard for embedding compression experiments
|
| 14 |
+
(binarization, MRL truncation, INT8/INT4, PQ, TurboQuant) on `BAAI/bge-base-en-v1.5`.
|
| 15 |
+
|
| 16 |
+
The page is fully client-side: `index.html` embeds the aggregate metrics and lazily
|
| 17 |
+
fetches per-subset / per-metric shards from `significance_shards/` and `per_query_shards/`.
|
| 18 |
+
|
| 19 |
+
Tabs: model comparison, size-vs-performance, full metrics table, per-query browser,
|
| 20 |
+
pairwise significance (Wilcoxon / paired t-test with normality diagnostics),
|
| 21 |
+
metric@K trend, and training trajectory.
|
index.html
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
per_query_shards/NanoArguAna.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
per_query_shards/NanoClimateFEVER.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
per_query_shards/NanoDBPedia.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ef4a87f6c63a6049f9f338453fd4d94e94a518cec47a4833609b301460ff06f3
|
| 3 |
+
size 19700584
|
per_query_shards/NanoFEVER.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
per_query_shards/NanoFiQA2018.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
per_query_shards/NanoHotpotQA.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
per_query_shards/NanoMSMARCO.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
per_query_shards/NanoNFCorpus.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3b66f2da78b830adceafd0742102ff7f4745e75ac75c45eafe315d2a08424e5
|
| 3 |
+
size 13968065
|
per_query_shards/NanoNQ.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
per_query_shards/NanoQuoraRetrieval.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
per_query_shards/NanoSCIDOCS.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9f48ea0e92466f3235ad9618a18d0d85597432b75d6d57fd19e5e26ead90a159
|
| 3 |
+
size 13474352
|
per_query_shards/NanoSciFact.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
per_query_shards/NanoTouche2020.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0594340409b622ef74d6e90bc11b4fda24c9d469e5035887d67fde2e1f437702
|
| 3 |
+
size 19973440
|
plotly.min.js
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__accuracy-1.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__accuracy-10.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__accuracy-3.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__accuracy-5.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__mrr-1.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__mrr-10.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__mrr-3.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__mrr-5.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__ndcg-1.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__ndcg-10.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__ndcg-3.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__ndcg-5.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__recall-1.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__recall-10.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__recall-3.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoArguAna__recall-5.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__accuracy-1.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__accuracy-10.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__accuracy-3.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__accuracy-5.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__mrr-1.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__mrr-10.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__mrr-3.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__mrr-5.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__ndcg-1.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__ndcg-10.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__ndcg-3.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__ndcg-5.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__recall-1.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__recall-10.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__recall-3.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoClimateFEVER__recall-5.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
significance_shards/NanoDBPedia__accuracy-1.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|