Labagaite commited on
Commit
778dce6
1 Parent(s): ad5e3ac

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +35 -5
README.md CHANGED
@@ -1,23 +1,53 @@
1
  ---
2
  language:
3
- - en
4
  license: apache-2.0
5
  tags:
6
  - text-generation-inference
7
  - transformers
8
  - unsloth
9
  - llama
10
- - trl
11
- - sft
12
  base_model: unsloth/llama-2-7b-chat-bnb-4bit
13
  ---
14
 
15
- # Uploaded model
16
 
17
  - **Developed by:** Labagaite
18
  - **License:** apache-2.0
19
  - **Finetuned from model :** unsloth/llama-2-7b-chat-bnb-4bit
20
 
21
- This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
 
 
 
 
 
 
 
 
 
 
22
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
23
  [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
 
 
 
1
  ---
2
  language:
3
+ - fr
4
  license: apache-2.0
5
  tags:
6
  - text-generation-inference
7
  - transformers
8
  - unsloth
9
  - llama
10
+ - summarizer
11
+ - 16bit
12
  base_model: unsloth/llama-2-7b-chat-bnb-4bit
13
  ---
14
 
15
+ # Uploaded as 16bit model
16
 
17
  - **Developed by:** Labagaite
18
  - **License:** apache-2.0
19
  - **Finetuned from model :** unsloth/llama-2-7b-chat-bnb-4bit
20
 
21
+ # Training Logs
22
+
23
+ ## Summary metrics
24
+ ### Best ROUGE-1 score : **0.007804878048780488**
25
+ ### Best ROUGE-2 score : **0**
26
+ ### Best ROUGE-L score : **0.005853658536585366**
27
+
28
+ ## Wandb logs
29
+ You can view the training logs [<img src="https://raw.githubusercontent.com/wandb/wandb/main/docs/README_images/logo-light.svg" width="200"/>](https://wandb.ai/william-derue/LLM-summarizer_trainer/runs/9wgdnwzd).
30
+
31
+ ## Training details
32
 
33
+ ### training data
34
+ - Dataset : [fr-summarizer-dataset](https://huggingface.co/datasets/Labagaite/fr-summarizer-dataset)
35
+ - Data-size : 7.65 MB
36
+ - train : 1.97k rows
37
+ - validation : 440 rows
38
+ - roles : user , assistant
39
+ - Format chatml "role": "role", "content": "content", "user": "user", "assistant": "assistant"
40
+ <br>
41
+ *French audio podcast transcription*
42
+
43
+ # Project details
44
+ [<img src="https://avatars.githubusercontent.com/u/116890814?v=4" width="100"/>](https://github.com/WillIsback/Report_Maker)
45
+ Fine-tuned on French audio podcast transcription data for summarization task. As a result, the model is able to summarize French audio podcast transcription data.
46
+ The model will be used for an AI application: [Report Maker](https://github.com/WillIsback/Report_Maker) wich is a powerful tool designed to automate the process of transcribing and summarizing meetings.
47
+ It leverages state-of-the-art machine learning models to provide detailed and accurate reports.
48
+
49
+ This llama model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
50
+ This llama was trained with [LLM summarizer trainer](images/Llm_Summarizer_trainer_icon-removebg.png)
51
  [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
52
+ **LLM summarizer trainer**
53
+ [<img src="images/Llm_Summarizer_trainer_icon-removebg.png" width="150"/>](https://github.com/WillIsback/LLM_Summarizer_Trainer)