Update README.md
Browse files
README.md
CHANGED
|
@@ -1,5 +1,10 @@
|
|
| 1 |
---
|
| 2 |
library_name: peft
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
| 4 |
|
| 5 |
# video-SALMONN 2+ (Qwen 2.5-VL Based video-SALMONN 2)
|
|
@@ -37,8 +42,6 @@ Other benchmarks
|
|
| 37 |
|
| 38 |
## How to Use
|
| 39 |
|
| 40 |
-
**Due to unknown BUGs, when using `from_pretrained` and `device_map="auto"`, the model generates weird result, which forces the code to use [ZeRO Inference](https://www.deepspeed.ai/2022/09/09/zero-inference.html) when testing the 72B model.**
|
| 41 |
-
|
| 42 |
1. Prepare the dataset following `scripts/example_av.json`, `scripts/example_v.json`, `scripts/example_dpo.json`, and `scripts/example_a.json`
|
| 43 |
2. Prepare base audio model through modifying the path in `gen_audio_model.py`
|
| 44 |
3. To conduct audio alignment, use the following script:
|
|
|
|
| 1 |
---
|
| 2 |
library_name: peft
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
language:
|
| 5 |
+
- en
|
| 6 |
+
base_model:
|
| 7 |
+
- Qwen/Qwen2.5-VL-7B-Instruct
|
| 8 |
---
|
| 9 |
|
| 10 |
# video-SALMONN 2+ (Qwen 2.5-VL Based video-SALMONN 2)
|
|
|
|
| 42 |
|
| 43 |
## How to Use
|
| 44 |
|
|
|
|
|
|
|
| 45 |
1. Prepare the dataset following `scripts/example_av.json`, `scripts/example_v.json`, `scripts/example_dpo.json`, and `scripts/example_a.json`
|
| 46 |
2. Prepare base audio model through modifying the path in `gen_audio_model.py`
|
| 47 |
3. To conduct audio alignment, use the following script:
|