File size: 1,928 Bytes
0f64ab8
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7c62e54
a0ac58e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
be9fc1c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
a0ac58e
 
 
 
 
 
7c62e54
 
 
 
708e9fb
7c62e54
708e9fb
7c62e54
a0ac58e
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
---

language:
- en
license: apache-2.0
tags:
- text-generation-inference
- transformers
- unsloth
- llama3
- trl
- hinglish
base_model: unsloth/llama-3-8b
datasets:
- cmu_hinglish_dog
---


# Inference:

```
!pip install -q "unsloth[colab-new] @ git+https://github.com/unslothai/unsloth.git"
!pip install -q --no-deps "xformers<0.0.26" trl peft accelerate bitsandbytes
```

```python
from unsloth import FastLanguageModel
import torch
max_seq_length = 512
dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+
load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False.

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "Hinglish-Project/llama-3-8b-English-to-Hinglish",
    max_seq_length = max_seq_length,
    dtype = dtype,
    load_in_4bit = load_in_4bit,
)
```

```python
def pipe(prompt):
  alpaca_prompt = """### Instrucion: Translate given text to Hinglish Text:

### Input:
{}

### Response:
"""

  inputs = tokenizer(
      [
          alpaca_prompt.format(prompt),
      ], return_tensors = "pt").to("cuda")

  outputs = model.generate(**inputs, max_new_tokens = 2048, use_cache = True)
  raw_text = tokenizer.batch_decode(outputs)[0]
  return raw_text.split("### Response:\n")[1].split("<|end_of_text|>")[0]
```

```python
text = "This is a fine-tuned Hinglish translation model using Llama 3."
pipe(text)
## yeh ek fine-tuned Hinglish translation model hai jisme Llama 3 ka use kiya gaya hai.

```
# Uploaded  model

- **Developed by:** Hinglish-Project
- **License:** apache-2.0
- **Finetuned from model :** unsloth/llama-8b-bnb-4bit

This Llama3 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.

[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)