Commit
•
30a9e3a
1
Parent(s):
f5f37bf
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
language:
|
4 |
+
- ta
|
5 |
+
library_name: transformers
|
6 |
+
pipeline_tag: text-generation
|
7 |
+
---
|
8 |
+
# Model Card for Model ID
|
9 |
+
|
10 |
+
<!-- Provide a quick summary of what the model is/does. -->
|
11 |
+
|
12 |
+
This model is trained on PonniyinSelvan tamil corpus dataset.
|
13 |
+
|
14 |
+
## Model Details
|
15 |
+
Base model used is EleutherAI's Pythia 1.4b
|
16 |
+
### Model Description
|
17 |
+
|
18 |
+
|
19 |
+
- **Finetuned from model [optional]:** Pythia 1.4b
|
20 |
+
|
21 |
+
|
22 |
+
## Uses
|
23 |
+
|
24 |
+
Purely education and research purposes only. Not fit for any kind of practical use.
|
25 |
+
|
26 |
+
|
27 |
+
|
28 |
+
## Bias, Risks, and Limitations
|
29 |
+
|
30 |
+
The base model Bias, Risks and Limitations apply
|
31 |
+
|
32 |
+
|
33 |
+
## How to Get Started with the Model
|
34 |
+
```python
|
35 |
+
import torch
|
36 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM
|
37 |
+
|
38 |
+
model_path = "RajuKandasamy/ponniyinselvan_1.4b_alpha"
|
39 |
+
device = "cuda" if torch.cuda.is_available() else "cpu"
|
40 |
+
model = AutoModelForCausalLM.from_pretrained(model_path, load_in_8bit=False).to(device)
|
41 |
+
|
42 |
+
tokenizer = AutoTokenizer.from_pretrained(model_path)
|
43 |
+
|
44 |
+
model.eval()
|
45 |
+
|
46 |
+
prompt="""வந்தியத்தேவன்"""
|
47 |
+
input_ids = tokenizer.encode(prompt, return_tensors="pt").to(model.device)
|
48 |
+
attention_mask = torch.ones_like(input_ids).to(model.device) # set attention mask to 1 for all input tokens
|
49 |
+
print("Thinking ...\n ")
|
50 |
+
with torch.no_grad():
|
51 |
+
output = model.generate(input_ids=input_ids, attention_mask=attention_mask, max_length=256, early_stopping=False, temperature=0.9, top_p=0.9,top_k=500, do_sample=True,output_scores=True, pad_token_id=tokenizer.eos_token_id, repetition_penalty=1.2,eos_token_id=tokenizer.eos_token_id)
|
52 |
+
output_str = tokenizer.decode(output[0], skip_special_tokens=False)
|
53 |
+
print(output_str)
|
54 |
+
```
|
55 |
+
|
56 |
+
## Training Details
|
57 |
+
|
58 |
+
10 epochs
|
59 |
+
|
60 |
+
### Training Data
|
61 |
+
|
62 |
+
ponniyinselvan text corpus
|
63 |
+
|
64 |
+
### Training Procedure
|
65 |
+
|
66 |
+
Casual Language Modelling, With custom BPE tokenizer
|
67 |
+
|
68 |
+
|