Commit
•
980a6e5
1
Parent(s):
df8d0b1
readme description added.
Browse files
README.md
ADDED
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
---
|
5 |
+
This is llama2 7B finetuned using qlora with bf16 as compute dtype. The dataset has been generated using open-ai api with samples semantics oriented towards abstract explanation of system design.
|
6 |
+
|
7 |
+
|
8 |
+
lora has been merged into the original model, 3 peochs have been trained with batch size of 16.
|
9 |
+
|
10 |
+
|
11 |
+
```bash
|
12 |
+
from google.colab import drive
|
13 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
14 |
+
from transformers import pipeline
|
15 |
+
|
16 |
+
model_path = "SaffalPoosh/system_design_expert"
|
17 |
+
|
18 |
+
model = AutoModelForCausalLM.from_pretrained(model_path)
|
19 |
+
tokenizer = AutoTokenizer.from_pretrained(model_path)
|
20 |
+
|
21 |
+
|
22 |
+
prompt = "Design an application like Whatsapp with tech stack you will use"
|
23 |
+
gen = pipeline('text-generation', model=model, tokenizer=tokenizer)
|
24 |
+
result = gen(prompt)
|
25 |
+
print(result[0]['generated_text'])
|
26 |
+
```
|
27 |
+
|