Update README.md
Browse files
README.md
CHANGED
|
@@ -58,7 +58,7 @@ model = AutoModelForCausalLM.from_pretrained("snorkelai/RedPajama-7B-Chat-Curate
|
|
| 58 |
model = model.to('cuda:0')
|
| 59 |
|
| 60 |
## inference
|
| 61 |
-
prompt = "<human>:
|
| 62 |
inputs = tokenizer(prompt, return_tensors='pt').to(model.device)
|
| 63 |
input_length = inputs.input_ids.shape[1]
|
| 64 |
outputs = model.generate(
|
|
@@ -67,7 +67,20 @@ outputs = model.generate(
|
|
| 67 |
token = outputs.sequences[0, input_length:]
|
| 68 |
output_str = tokenizer.decode(token)
|
| 69 |
print(output_str)
|
| 70 |
-
"""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 71 |
```
|
| 72 |
|
| 73 |
Comparing with RedPajama-INCITE-7B-Chat
|
|
@@ -78,7 +91,7 @@ model = AutoModelForCausalLM.from_pretrained("togethercomputer/RedPajama-INCITE-
|
|
| 78 |
model = model.to('cuda:0')
|
| 79 |
|
| 80 |
## inference
|
| 81 |
-
prompt = "<human>:
|
| 82 |
inputs = tokenizer(prompt, return_tensors='pt').to(model.device)
|
| 83 |
input_length = inputs.input_ids.shape[1]
|
| 84 |
outputs = model.generate(
|
|
@@ -87,9 +100,20 @@ outputs = model.generate(
|
|
| 87 |
token = outputs.sequences[0, input_length:]
|
| 88 |
output_str = tokenizer.decode(token)
|
| 89 |
print(output_str)
|
| 90 |
-
"""{EDIT_NEEDED}"""
|
| 91 |
"""
|
| 92 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 93 |
"""
|
| 94 |
```
|
| 95 |
|
|
|
|
| 58 |
model = model.to('cuda:0')
|
| 59 |
|
| 60 |
## inference
|
| 61 |
+
prompt = "<human>: Give me a list of food recommendations to try in Japan.\n<bot>:"
|
| 62 |
inputs = tokenizer(prompt, return_tensors='pt').to(model.device)
|
| 63 |
input_length = inputs.input_ids.shape[1]
|
| 64 |
outputs = model.generate(
|
|
|
|
| 67 |
token = outputs.sequences[0, input_length:]
|
| 68 |
output_str = tokenizer.decode(token)
|
| 69 |
print(output_str)
|
| 70 |
+
"""
|
| 71 |
+
Here are some food recommendations to try in Japan:
|
| 72 |
+
|
| 73 |
+
1. Sushi: Japan is famous for its sushi, a dish of vinegared rice with various ingredients such as seafood, vegetables, and meat.
|
| 74 |
+
2. Tempura: Another popular Japanese dish, tempura is a vegetable or seafood batter that is deep-fried.
|
| 75 |
+
3. Yakitori: Grilled chicken skewers are a common street food in Japan.
|
| 76 |
+
4. Ramen: A noodle soup dish that is a staple food in Japan.
|
| 77 |
+
5. Bentos: A Japanese boxed lunch that is a convenient and affordable option for travelers.
|
| 78 |
+
6. Fugu: A type of fish that contains highly poisonous pufferfish. It is considered a delicacy in Japan and is only prepared by trained professionals.
|
| 79 |
+
7. Sukiyaki: A Japanese hot pot dish made with beef and vegetables.
|
| 80 |
+
8. Takoyaki: A Japanese dumpling dish filled with octopus and pickled ginger.
|
| 81 |
+
9. Gyoza: Chinese-style dumplings that are a popular street food in Japan.
|
| 82 |
+
10. Matcha: A type of green tea that is often served as a latte or in desserts.
|
| 83 |
+
"""
|
| 84 |
```
|
| 85 |
|
| 86 |
Comparing with RedPajama-INCITE-7B-Chat
|
|
|
|
| 91 |
model = model.to('cuda:0')
|
| 92 |
|
| 93 |
## inference
|
| 94 |
+
prompt = "<human>: Give me a list of food recommendations to try in Japan.\n<bot>:"
|
| 95 |
inputs = tokenizer(prompt, return_tensors='pt').to(model.device)
|
| 96 |
input_length = inputs.input_ids.shape[1]
|
| 97 |
outputs = model.generate(
|
|
|
|
| 100 |
token = outputs.sequences[0, input_length:]
|
| 101 |
output_str = tokenizer.decode(token)
|
| 102 |
print(output_str)
|
|
|
|
| 103 |
"""
|
| 104 |
+
- Sushi
|
| 105 |
+
- Ramen
|
| 106 |
+
- Tempura
|
| 107 |
+
- Yakitori
|
| 108 |
+
- Sujuki
|
| 109 |
+
- Gyoza
|
| 110 |
+
- Takoyaki
|
| 111 |
+
- Okonomiyaki
|
| 112 |
+
- Baccala Tokkumon
|
| 113 |
+
- Takoyaki Mayo
|
| 114 |
+
- Karaage
|
| 115 |
+
- Gohan bowl
|
| 116 |
+
- California Maki
|
| 117 |
"""
|
| 118 |
```
|
| 119 |
|