Files
wehub-resource-sync a203934033
Lint test / lint (push) Has been cancelled
chore: import upstream snapshot with attribution
2026-07-13 13:34:58 +08:00

169 lines
5.9 KiB
Python
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
from swift.model import get_processor
from swift.template import TemplateInputs, get_template
def test_deepseek_v2_5():
tokenizer = get_processor('deepseek-ai/DeepSeek-V2.5-1210')
template = get_template(tokenizer)
inputs = TemplateInputs({
'messages': [{
'role': 'system',
'content': '000'
}, {
'role': 'user',
'content': 'aaa'
}, {
'role': 'assistant',
'content': 'bbb'
}, {
'role': 'user',
'content': 'ccc'
}]
})
res = template.encode(inputs)
template.print_inputs(res)
template.template_backend = 'jinja'
res2 = template.encode(inputs)
template.print_inputs(res2)
assert res['input_ids'] == res2['input_ids']
def test_qwen2_5_math_reward():
tokenizer = get_processor('Qwen/Qwen2.5-Math-RM-72B')
template = get_template(tokenizer)
inputs = TemplateInputs({
'messages': [{
'role':
'user',
'content':
'Janets ducks lay 16 eggs per day. She eats three for breakfast every morning and bakes muffins '
"for her friends every day with four. She sells the remainder at the farmers' market daily for $2 per "
"fresh duck egg. How much in dollars does she make every day at the farmers' market?"
}, {
'role':
'assistant',
'content':
"To determine how much Janet makes from selling the duck eggs at the farmers' market, we need to "
'follow these steps:\n\n1. Calculate the total number of eggs laid by the ducks each day.\n2. '
'Determine how many eggs Janet eats and bakes for herself each day.\n3. Find out how many eggs are '
"left to be sold.\n4. Calculate the revenue from selling the remaining eggs at $2 per egg.\n\nLet's "
"start with the first step:\n\n1. Janet's ducks lay 16 eggs per day.\n\nNext, we calculate how many "
'eggs Janet eats and bakes for herself each day:\n\n2. Janet eats 3 eggs for breakfast every morning.'
'\n3. Janet bakes 4 eggs for her friends every day.\n\nSo, the total number of eggs Janet eats and '
'bakes for herself each day is:\n\\[ 3 + 4 = 7 \\text{ eggs} \\]\n\nNow, we find out how many eggs '
'are left to be sold:\n\\[ 16 - 7 = 9 \\text{ eggs} \\]\n\nFinally, we calculate the revenue from '
'selling the remaining eggs at $2 per egg:\n\\[ 9 \\times 2 = 18 \\text{ dollars} \\]\n\nTherefore, '
"Janet makes \\(\\boxed{18}\\) dollars every day at the farmers' market."
}]
})
res = template.encode(inputs)
template.print_inputs(res)
template.template_backend = 'jinja'
res2 = template.encode(inputs)
template.print_inputs(res)
assert res['input_ids'] == res2['input_ids']
assert len(res['input_ids']) == 364
def test_minimax():
tokenizer = get_processor('MiniMax/MiniMax-Text-01')
template = get_template(tokenizer)
inputs = TemplateInputs({
'messages': [{
'role': 'system',
'content': 'You are a helpful assistant created by MiniMax based on MiniMax-Text-01 model.'
}, {
'role': 'user',
'content': 'Hello!'
}]
})
res = template.encode(inputs)
template.print_inputs(res)
assert tokenizer.decode(res['input_ids']) == (
'<beginning_of_sentence>system ai_setting=assistant\nYou are a helpful assistant created by MiniMax based '
'on MiniMax-Text-01 model.<end_of_sentence>\n<beginning_of_sentence>user name=user\nHello!<end_of_sentence>\n'
'<beginning_of_sentence>ai name=assistant\n')
def test_minimax_vl():
tokenizer = get_processor('MiniMax/MiniMax-VL-01')
template = get_template(tokenizer)
inputs = TemplateInputs({
'messages': [{
'role': 'system',
'content': 'You are a helpful assistant created by MiniMax based on MiniMax-VL-01 model.'
}, {
'role': 'user',
'content': '<image>Describe this image.'
}],
'images': ['http://modelscope-open.oss-cn-hangzhou.aliyuncs.com/images/cat.png']
})
res = template.encode(inputs)
assert len(res['input_ids']) == 5877
def test_deepseek_v3_1():
tokenizer = get_processor('deepseek-ai/DeepSeek-V3.1')
template = get_template(tokenizer)
inputs = {
'messages': [{
'role': 'system',
'content': '000'
}, {
'role': 'user',
'content': 'aaa'
}, {
'role': 'assistant',
'content': 'bbb'
}, {
'role': 'user',
'content': 'ccc'
}]
}
res = template.encode(inputs)
template.print_inputs(res)
template.template_backend = 'jinja'
res2 = template.encode(inputs)
template.print_inputs(res2)
assert res['input_ids'] == res2['input_ids']
def test_preserve_thinking():
tokenizer = get_processor('Qwen/Qwen3.6-35B-A3B')
template = get_template(tokenizer, preserve_thinking=True)
template.set_mode('train')
inputs = {
'messages': [{
'role': 'system',
'content': '000'
}, {
'role': 'user',
'content': 'aaa'
}, {
'role': 'assistant',
'content': '<think>\nbbb\n</think>\n\nbbb'
}, {
'role': 'user',
'content': 'ccc'
}, {
'role': 'assistant',
'content': '<think>\nddd\n</think>\n\nddd'
}]
}
template.template_backend = 'swift'
res = template.encode(inputs)
template.print_inputs(res)
template.template_backend = 'jinja'
res2 = template.encode(inputs)
template.print_inputs(res2)
assert res['input_ids'] == res2['input_ids']
if __name__ == '__main__':
# test_deepseek_v2_5()
# test_qwen2_5_math_reward()
# test_minimax()
# test_minimax_vl()
# test_deepseek_v3_1()
test_preserve_thinking()