kurianbenoy commited on
Commit
c72bdac
1 Parent(s): 5426eef

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +12 -15
README.md CHANGED
@@ -3,30 +3,27 @@ tags:
3
  - fastai
4
  ---
5
 
6
- # Amazing!
7
 
8
- 🥳 Congratulations on hosting your fastai model on the Hugging Face Hub!
9
 
10
- # Some next steps
11
- 1. Fill out this model card with more information (see the template below and the [documentation here](https://huggingface.co/docs/hub/model-repos))!
12
 
13
- 2. Create a demo in Gradio or Streamlit using 🤗 Spaces ([documentation here](https://huggingface.co/docs/hub/spaces)).
14
 
15
- 3. Join the fastai community on the [Fastai Discord](https://discord.com/invite/YKrxeNn)!
16
 
17
- Greetings fellow fastlearner 🤝! Don't forget to delete this content from your model card.
18
 
 
19
 
20
- ---
21
 
 
 
22
 
23
- # Model card
24
 
25
- ## Model description
26
- More information needed
27
-
28
- ## Intended uses & limitations
29
- More information needed
30
 
31
  ## Training and evaluation data
32
- More information needed
 
 
 
3
  - fastai
4
  ---
5
 
 
6
 
7
+ # Model card
8
 
9
+ ## Model description
 
10
 
11
+ This model has been trained with convnext_tiny_in22k with [Flowers-101 datasets in Kaggle](https://www.kaggle.com/competitions/tpu-getting-started).
12
 
 
13
 
14
+ ## Intended uses & limitations
15
 
16
+ - The model can be used be for classifying flowers only.
17
 
18
+ **Limitations**
19
 
20
+ - Even if the picture uploaded is not of a flower, you can can notice [it will be predicted as of flower](https://www.kaggle.com/competitions/tpu-getting-started).
21
+ - The model on validation dataset has accuracy of 94.23%
22
 
23
+ ![image](https://user-images.githubusercontent.com/24592806/177065484-1fae6a79-5dbe-471a-8c86-9c5aaa336bc6.png)
24
 
 
 
 
 
 
25
 
26
  ## Training and evaluation data
27
+
28
+ - The models has been trained and evaluated with [Flowers-101 datasets in Kaggle](https://www.kaggle.com/competitions/tpu-getting-started).
29
+ - We used a Random Splitter to train and evaluate data