Hugging Face's logo

shorecode
/
t5-efficient-tiny-summarizer-general-purpose-v2

Transformers
PyTorch
TensorFlow
ONNX
Safetensors
English
t5
text2text-generation
summary
summarizer
Eval Results (legacy)
text-generation-inference
Model card Files
xet
Community
shorecode commited on
Commit
871a3b8
·
verified ·
1 Parent(s): 603e4b9

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +35 -8
README.md CHANGED
@@ -28,7 +28,7 @@ model-index:
28
  metrics:
29
  - name: f1 Score
30
  type: f1 Score
31
- value: 0.308
32
  - task:
33
  type: Summarization
34
  dataset:
@@ -37,7 +37,7 @@ model-index:
37
  metrics:
38
  - name: Faithfullness (facebook/bart-large-cnn)
39
  type: facebook/bart-large-cnn
40
- value: 2.91
41
  - task:
42
  type: Summarization
43
  dataset:
@@ -46,7 +46,7 @@ model-index:
46
  metrics:
47
  - name: Summarization Compression
48
  type: Lighteval extractiveness
49
- value: 7.48
50
  - task:
51
  type: Summarization
52
  dataset:
@@ -55,7 +55,7 @@ model-index:
55
  metrics:
56
  - name: Summarization Coverage
57
  type: Lighteval extractiveness
58
- value: 0.90
59
  - task:
60
  type: Summarization
61
  dataset:
@@ -64,7 +64,7 @@ model-index:
64
  metrics:
65
  - name: Summarization Density
66
  type: Lighteval extractiveness
67
- value: 5.78
68
  - task:
69
  type: Summarization
70
  dataset:
@@ -73,7 +73,7 @@ model-index:
73
  metrics:
74
  - name: rougeL precision
75
  type: Lighteval
76
- value: 0.56
77
  - task:
78
  type: Summarization
79
  dataset:
@@ -82,7 +82,7 @@ model-index:
82
  metrics:
83
  - name: rougeL recall
84
  type: Lighteval
85
- value: 0.12
86
  - task:
87
  type: Summarization
88
  dataset:
@@ -91,7 +91,34 @@ model-index:
91
  metrics:
92
  - name: rougeL fmeasure
93
  type: Lighteval
94
- value: 0.19
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
95
  ---
96
 
97
  # This model was built to shorten text to be injected into LLM prompts to reduce API calling costs
 
28
  metrics:
29
  - name: f1 Score
30
  type: f1 Score
31
+ value: 0.323
32
  - task:
33
  type: Summarization
34
  dataset:
 
37
  metrics:
38
  - name: Faithfullness (facebook/bart-large-cnn)
39
  type: facebook/bart-large-cnn
40
+ value: 2.56
41
  - task:
42
  type: Summarization
43
  dataset:
 
46
  metrics:
47
  - name: Summarization Compression
48
  type: Lighteval extractiveness
49
+ value: 6.91
50
  - task:
51
  type: Summarization
52
  dataset:
 
55
  metrics:
56
  - name: Summarization Coverage
57
  type: Lighteval extractiveness
58
+ value: 0.89
59
  - task:
60
  type: Summarization
61
  dataset:
 
64
  metrics:
65
  - name: Summarization Density
66
  type: Lighteval extractiveness
67
+ value: 4.95
68
  - task:
69
  type: Summarization
70
  dataset:
 
73
  metrics:
74
  - name: rougeL precision
75
  type: Lighteval
76
+ value: 0.51
77
  - task:
78
  type: Summarization
79
  dataset:
 
82
  metrics:
83
  - name: rougeL recall
84
  type: Lighteval
85
+ value: 0.13
86
  - task:
87
  type: Summarization
88
  dataset:
 
91
  metrics:
92
  - name: rougeL fmeasure
93
  type: Lighteval
94
+ value: 0.20
95
+ - task:
96
+ type: Summarization
97
+ dataset:
98
+ name: shorecode/summary-collection-60k-rows
99
+ type: shorecode/summary-collection-60k-rows
100
+ metrics:
101
+ - name: rouge1 precision
102
+ type: Lighteval
103
+ value: 0.71
104
+ - task:
105
+ type: Summarization
106
+ dataset:
107
+ name: shorecode/summary-collection-60k-rows
108
+ type: shorecode/summary-collection-60k-rows
109
+ metrics:
110
+ - name: rouge1 recall
111
+ type: Lighteval
112
+ value: 0.18
113
+ - task:
114
+ type: Summarization
115
+ dataset:
116
+ name: shorecode/summary-collection-60k-rows
117
+ type: shorecode/summary-collection-60k-rows
118
+ metrics:
119
+ - name: rouge1 fmeasure
120
+ type: Lighteval
121
+ value: 0.28
122
  ---
123
 
124
  # This model was built to shorten text to be injected into LLM prompts to reduce API calling costs