offlinedemo 0.3.0__tar.gz → 0.4.0__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {offlinedemo-0.3.0 → offlinedemo-0.4.0}/PKG-INFO +5 -2
- {offlinedemo-0.3.0 → offlinedemo-0.4.0}/README.md +4 -1
- {offlinedemo-0.3.0 → offlinedemo-0.4.0}/pyproject.toml +1 -1
- {offlinedemo-0.3.0 → offlinedemo-0.4.0}/pyproject.toml.orig +1 -1
- {offlinedemo-0.3.0 → offlinedemo-0.4.0}/src/offlinedemo/__init__.py +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.3
|
|
2
2
|
Name: offlinedemo
|
|
3
|
-
Version: 0.
|
|
3
|
+
Version: 0.4.0
|
|
4
4
|
Summary: Easy demonstration of models by offlineisbetter.
|
|
5
5
|
Author: offlineisbetter
|
|
6
6
|
Author-email: offlineisbetter <dev@offlineisbetter.com>
|
|
@@ -95,11 +95,14 @@ offlinedemo offline-sentiment-small
|
|
|
95
95
|
|
|
96
96
|
## benchmarks
|
|
97
97
|
|
|
98
|
+
below are benchmarks for `offline-sentiment-small` on binary sentiment classification using the [stanfordnlp/sst2](https://huggingface.co/datasets/stanfordnlp/sst2) validation set. all benchmarks were completed on the ryzen 9950x3d cpu on one thread.
|
|
99
|
+
|
|
98
100
|
| model | parameters | p95 (ms) | f1 (validation) |
|
|
99
101
|
|:---:|:---:|:---:|:---:|
|
|
100
102
|
| `offline-sentiment-small` | 230m | 80.32 | 0.9489 |
|
|
101
103
|
| `distilbert-base` | 67m | 66.15 | 0.9321 |
|
|
102
|
-
| `roberta-base` |
|
|
104
|
+
| `roberta-base` | 125m | 469.65 | 0.9396 |
|
|
105
|
+
| `modernbert-base` | 149m | 530.73 | 0.9396 |
|
|
103
106
|
|
|
104
107
|
## philosophy
|
|
105
108
|
|
|
@@ -28,11 +28,14 @@ offlinedemo offline-sentiment-small
|
|
|
28
28
|
|
|
29
29
|
## benchmarks
|
|
30
30
|
|
|
31
|
+
below are benchmarks for `offline-sentiment-small` on binary sentiment classification using the [stanfordnlp/sst2](https://huggingface.co/datasets/stanfordnlp/sst2) validation set. all benchmarks were completed on the ryzen 9950x3d cpu on one thread.
|
|
32
|
+
|
|
31
33
|
| model | parameters | p95 (ms) | f1 (validation) |
|
|
32
34
|
|:---:|:---:|:---:|:---:|
|
|
33
35
|
| `offline-sentiment-small` | 230m | 80.32 | 0.9489 |
|
|
34
36
|
| `distilbert-base` | 67m | 66.15 | 0.9321 |
|
|
35
|
-
| `roberta-base` |
|
|
37
|
+
| `roberta-base` | 125m | 469.65 | 0.9396 |
|
|
38
|
+
| `modernbert-base` | 149m | 530.73 | 0.9396 |
|
|
36
39
|
|
|
37
40
|
## philosophy
|
|
38
41
|
|
|
File without changes
|