Hugging Face's logo

Datasets:
happy8825
/
best_eval

Modalities:
Tabular
Text
Formats:
json
Languages:
English
Size:
< 1K
Tags:
video-retrieval
evaluation
vllm
Libraries:
Datasets
pandas
Polars
Dataset card Data Studio Files
xet
Community
1
happy8825 commited on
Commit
f2a1793
·
verified ·
1 Parent(s): 30ebae5

Update README with metrics

Browse files
Files changed (1) hide show
  1. README.md +6 -6
README.md CHANGED
@@ -1,5 +1,5 @@
1
  ---
2
- pretty_name: "/hub_data4/seohyun/saves/ecva_instruct_1223/full/sft/checkpoint-350/ · happy8825/valid_ecva_clean results"
3
  language:
4
  - en
5
  tags:
@@ -8,11 +8,11 @@ tags:
8
  - vllm
9
  ---
10
 
11
- # /hub_data4/seohyun/saves/ecva_instruct_1223/full/sft/checkpoint-350/ · happy8825/valid_ecva_clean results
12
 
13
- - **Model**: `/hub_data4/seohyun/saves/ecva_instruct_1223/full/sft/checkpoint-350/`
14
  - **Dataset**: `happy8825/valid_ecva_clean`
15
- - **Generated**: `2025-12-24 03:23:22Z`
16
 
17
  ## Metrics
18
  | Metric | Value |
@@ -31,7 +31,7 @@ The uploaded JSON contains full per-sample predictions produced via `t3_infer_wi
31
  | --- | --- |
32
  | EVQA total | 924 |
33
  | EVQA with GT label | 924 |
34
- | EVQA accuracy | 0.756494 |
35
 
36
  ## Run Summary
37
 
@@ -48,7 +48,7 @@ Metrics: {
48
  "shard_index": 0,
49
  "evqa_total": 924,
50
  "evqa_with_gt_label": 924,
51
- "evqa_acc": 0.7564935064935064
52
  }
53
  Pushed best_eval.jsonl and README to https://huggingface.co/datasets/happy8825/best_eval
54
  ```
 
1
  ---
2
+ pretty_name: "/hub_data4/seohyun/saves/ecva_instruct_1223/full/sft/checkpoint-450/ · happy8825/valid_ecva_clean results"
3
  language:
4
  - en
5
  tags:
 
8
  - vllm
9
  ---
10
 
11
+ # /hub_data4/seohyun/saves/ecva_instruct_1223/full/sft/checkpoint-450/ · happy8825/valid_ecva_clean results
12
 
13
+ - **Model**: `/hub_data4/seohyun/saves/ecva_instruct_1223/full/sft/checkpoint-450/`
14
  - **Dataset**: `happy8825/valid_ecva_clean`
15
+ - **Generated**: `2025-12-24 03:42:34Z`
16
 
17
  ## Metrics
18
  | Metric | Value |
 
31
  | --- | --- |
32
  | EVQA total | 924 |
33
  | EVQA with GT label | 924 |
34
+ | EVQA accuracy | 0.520563 |
35
 
36
  ## Run Summary
37
 
 
48
  "shard_index": 0,
49
  "evqa_total": 924,
50
  "evqa_with_gt_label": 924,
51
+ "evqa_acc": 0.5205627705627706
52
  }
53
  Pushed best_eval.jsonl and README to https://huggingface.co/datasets/happy8825/best_eval
54
  ```