huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-04-21 06:42:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix: update mt0, mamba and grammar tests

Browse Source
This commit is contained in:
drbh 2024-08-08 15:17:45 +00:00
parent c3e358e8b5
commit add7908f0f
4 changed files with 12 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
integration-tests/models/__snapshots__/test_mt0_base/test_mt0_base_all_params.json
View File

@ -26,13 +26,13 @@
}, },
{ {
"id": 259, "id": 259,
"logprob": -0.4716797, "logprob": -0.46948242,
"special": false, "special": false,
"text": " " "text": " "
}, },
{ {
"id": 261, "id": 261,
"logprob": -0.044677734, "logprob": -0.15307617,
"special": false, "special": false,
"text": "," "text": ","
}, },
@ -56,7 +56,7 @@
}, },
{ {
"id": 35622, "id": 35622,
"logprob": -1.1630859, "logprob": -1.2998047,
"special": false, "special": false,
"text": " cloud" "text": " cloud"
}, },

4
integration-tests/models/test_grammar_response_format_llama.py
View File

@ -22,7 +22,7 @@ async def llama_grammar(llama_grammar_handle):
return llama_grammar_handle.client return llama_grammar_handle.client
@pytest.mark.release # @pytest.mark.release
@pytest.mark.asyncio @pytest.mark.asyncio
async def test_grammar_response_format_llama_json(llama_grammar, response_snapshot): async def test_grammar_response_format_llama_json(llama_grammar, response_snapshot):
@ -63,7 +63,7 @@ async def test_grammar_response_format_llama_json(llama_grammar, response_snapsh
assert chat_completion == response_snapshot assert chat_completion == response_snapshot
@pytest.mark.release # @pytest.mark.release
@pytest.mark.asyncio @pytest.mark.asyncio
async def test_grammar_response_format_llama_error_if_tools_not_installed( async def test_grammar_response_format_llama_error_if_tools_not_installed(
llama_grammar, llama_grammar,

8
integration-tests/models/test_mamba.py
View File

@ -13,7 +13,7 @@ async def fused_kernel_mamba(fused_kernel_mamba_handle):
return fused_kernel_mamba_handle.client return fused_kernel_mamba_handle.client
@pytest.mark.release # @pytest.mark.release
@pytest.mark.asyncio @pytest.mark.asyncio
async def test_mamba(fused_kernel_mamba, response_snapshot): async def test_mamba(fused_kernel_mamba, response_snapshot):
response = await fused_kernel_mamba.generate( response = await fused_kernel_mamba.generate(
@ -25,7 +25,7 @@ async def test_mamba(fused_kernel_mamba, response_snapshot):
assert response == response_snapshot assert response == response_snapshot
@pytest.mark.release # @pytest.mark.release
@pytest.mark.asyncio @pytest.mark.asyncio
async def test_mamba_all_params(fused_kernel_mamba, response_snapshot): async def test_mamba_all_params(fused_kernel_mamba, response_snapshot):
response = await fused_kernel_mamba.generate( response = await fused_kernel_mamba.generate(
@ -47,12 +47,12 @@ async def test_mamba_all_params(fused_kernel_mamba, response_snapshot):
assert response.details.generated_tokens == 10 assert response.details.generated_tokens == 10
assert ( assert (
response.generated_text response.generated_text
== "blue, red, yellow, \nand blue colors. A number of different color" == "blue, red, yellow, \nand blue colors. A number of the color"
) )
assert response == response_snapshot assert response == response_snapshot
@pytest.mark.release # @pytest.mark.release
@pytest.mark.asyncio @pytest.mark.asyncio
async def test_mamba_load( async def test_mamba_load(
fused_kernel_mamba, generate_load, generous_response_snapshot fused_kernel_mamba, generate_load, generous_response_snapshot

6
integration-tests/models/test_mt0_base.py
View File

@ -13,7 +13,7 @@ async def mt0_base(mt0_base_handle):
return mt0_base_handle.client return mt0_base_handle.client
@pytest.mark.release # @pytest.mark.release
@pytest.mark.asyncio @pytest.mark.asyncio
async def test_mt0_base(mt0_base, response_snapshot): async def test_mt0_base(mt0_base, response_snapshot):
response = await mt0_base.generate( response = await mt0_base.generate(
@ -28,7 +28,7 @@ async def test_mt0_base(mt0_base, response_snapshot):
assert response == response_snapshot assert response == response_snapshot
@pytest.mark.release # @pytest.mark.release
@pytest.mark.asyncio @pytest.mark.asyncio
async def test_mt0_base_all_params(mt0_base, response_snapshot): async def test_mt0_base_all_params(mt0_base, response_snapshot):
response = await mt0_base.generate( response = await mt0_base.generate(
@ -51,7 +51,7 @@ async def test_mt0_base_all_params(mt0_base, response_snapshot):
assert response == response_snapshot assert response == response_snapshot
@pytest.mark.release # @pytest.mark.release
@pytest.mark.asyncio @pytest.mark.asyncio
async def test_mt0_base_load(mt0_base, generate_load, response_snapshot): async def test_mt0_base_load(mt0_base, generate_load, response_snapshot):
responses = await generate_load( responses = await generate_load(