mirrors/docling
1
0
Fork 0
mirror of https://github.com/DS4SD/docling.git synced 2025-07-27 04:24:45 +00:00
Code Issues Packages Projects Releases Wiki Activity

bumped the glm version and adjusted the tests

Browse Source 
Signed-off-by: Peter Staar <taa@zurich.ibm.com>
This commit is contained in:
Peter Staar 2024-09-17 13:53:05 +02:00
parent 30a0ef69b4
commit 39a8ecfd4c
10 changed files with 37 additions and 37 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

44
poetry.lock generated
View File

@ -854,35 +854,27 @@ files = [
[[package]]
name = "deepsearch-glm"
version = "0.21.0"
version = "0.21.1"
description = "Graph Language Models"
optional = false
python-versions = "<4.0,>=3.8"
files = [
{file = "deepsearch_glm-0.21.0-cp310-cp310-macosx_13_0_arm64.whl", hash = "sha256:a07f9ee8b9532f2f02ce363fefd4622178552032e2de8e4f540cab16852b3d6d"},
{file = "deepsearch_glm-0.21.0-cp310-cp310-macosx_13_0_x86_64.whl", hash = "sha256:27561630487dc331c30238f94b5f00b4e5e1359bcd120ee7e5d9f9e3b4d824a1"},
{file = "deepsearch_glm-0.21.0-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:05bfde0f2bea6f235deb66093c1553248374914bc793f6f1823e632d2da1a625"},
{file = "deepsearch_glm-0.21.0-cp310-cp310-macosx_14_0_x86_64.whl", hash = "sha256:ff5a7aa3ed12b1ad8f8cb290851d4ddeb87d3f486ac9a1e90f13d69ff40233e1"},
{file = "deepsearch_glm-0.21.0-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4e721314ee7b9d7ce9303d38f4038e33758a55b004691e32a5821818c1f42aff"},
{file = "deepsearch_glm-0.21.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2605a90146336f9cff2659d90f3dc9eea52a08b72ef8da211323b197ae61c557"},
{file = "deepsearch_glm-0.21.0-cp311-cp311-macosx_13_0_arm64.whl", hash = "sha256:7fc8c07002ff8ade6deb1dc6d6bd3d07371433aa242cbc02c20349764b23269a"},
{file = "deepsearch_glm-0.21.0-cp311-cp311-macosx_13_0_x86_64.whl", hash = "sha256:ba0cf927f4e9f2553e94349e29c07b4505b94deafa55cb65a19457ff83b8dc9e"},
{file = "deepsearch_glm-0.21.0-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:e69001e9b64ea5d45fd3e03ec7612f531ebcb0eee6f574cbe4976598d78ede3c"},
{file = "deepsearch_glm-0.21.0-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:f8846f9228065a89a0438453b067815f7ac28753217912b944d28ca0d68fae6a"},
{file = "deepsearch_glm-0.21.0-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:1a7dcd0fe3911df1821e343946e24443126d3cf6e7b6c13d7dfdd437fdaf013a"},
{file = "deepsearch_glm-0.21.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:86d97bc0c19672bee7723153b143f3e6f65d97497a22b4da19488aab5224f77f"},
{file = "deepsearch_glm-0.21.0-cp312-cp312-macosx_13_0_arm64.whl", hash = "sha256:d74caa7336d256c8091af4db99f2d5bd6f7cd9c9448b9a210e5512de67eaef54"},
{file = "deepsearch_glm-0.21.0-cp312-cp312-macosx_13_0_x86_64.whl", hash = "sha256:82a32f996f398425d62de3681ff7568ef18b9e0a8900c52ac9a780341f169073"},
{file = "deepsearch_glm-0.21.0-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:4c823e59e934716ccd93df309a37d56a7e75cffac9831cffb9a9a560b84feadc"},
{file = "deepsearch_glm-0.21.0-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:b2c897bc2d57e0d21c86fe5706b5458d2a948e7016b26730f4e04a892f12d690"},
{file = "deepsearch_glm-0.21.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4273fbda4cb25e949776ff81f60dde2ae278cab94a67babe7fab024f98dce993"},
{file = "deepsearch_glm-0.21.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:117b0556e2a36582590d5627582265498bff66dae74658a67a02ef3a76956e34"},
{file = "deepsearch_glm-0.21.0-cp39-cp39-macosx_13_0_arm64.whl", hash = "sha256:2ba54efd3661d7196f0a9828ebf642cb21bc2bd0594915e0486bd50b2ec0632c"},
{file = "deepsearch_glm-0.21.0-cp39-cp39-macosx_13_0_x86_64.whl", hash = "sha256:197ca6dc60330ff90c90dc85507899307353a2a0620db40ee825e632644c99c5"},
{file = "deepsearch_glm-0.21.0-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:a49c63cef72c32b63a10fc85589721bf82580a8a42cfe0a5c901798f4d161fbf"},
{file = "deepsearch_glm-0.21.0-cp39-cp39-macosx_14_0_x86_64.whl", hash = "sha256:3802436d610c6e24a0ea9291a2cdee9c260fd4492438af08041ed460e3f92743"},
{file = "deepsearch_glm-0.21.0-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:c1dd273c950877fb40d538ba16724efdedb82f3c9f15f9fc4407b9d60a832490"},
{file = "deepsearch_glm-0.21.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b9f0b0cbd2f773558f79d356603f1d49d913e52c8f9b5610b4603480df3c5804"},
{file = "deepsearch_glm-0.21.1-cp310-cp310-macosx_13_0_arm64.whl", hash = "sha256:b765d371ab0a4f57dd2532c651d7dc1b4a187395153e619a77b6f0d0f6aefb32"},
{file = "deepsearch_glm-0.21.1-cp310-cp310-macosx_13_0_x86_64.whl", hash = "sha256:c69e055b98d0a22267a1d0b6139801aecc5b7386289b89f53f976ab723352728"},
{file = "deepsearch_glm-0.21.1-cp310-cp310-macosx_14_0_arm64.whl", hash = "sha256:3eaa245e5ac4ab3e9d0c95a93e23f58d61d70f11431b76b6705fae358eb31c62"},
{file = "deepsearch_glm-0.21.1-cp310-cp310-macosx_14_0_x86_64.whl", hash = "sha256:63d195f6c5b30f4f908436589cffd4a5b9e18553c44c57fb635068a2afbd7fab"},
{file = "deepsearch_glm-0.21.1-cp311-cp311-macosx_13_0_arm64.whl", hash = "sha256:13bea2b4e8c04647ec743c3feb1ee66c784db542ab9dbed8dad7eb66fca74b70"},
{file = "deepsearch_glm-0.21.1-cp311-cp311-macosx_13_0_x86_64.whl", hash = "sha256:c5b8b8e2207615ff99e535f00548c7b0b8e4ca4593e59edd83fcad98fc318284"},
{file = "deepsearch_glm-0.21.1-cp311-cp311-macosx_14_0_arm64.whl", hash = "sha256:ba74868243caf5ac850fff7c45c8a372c1cac0193431e22eb41888d45ac79719"},
{file = "deepsearch_glm-0.21.1-cp311-cp311-macosx_14_0_x86_64.whl", hash = "sha256:7815b06aa1c3953488496f191ce0265d0ee7bed5a6b96454a5f9d6f1add28f69"},
{file = "deepsearch_glm-0.21.1-cp312-cp312-macosx_13_0_arm64.whl", hash = "sha256:fd4d0d4ff853e566b05769c704a4ea3c050c0cfc5721e4e2035e550fb2a8fe91"},
{file = "deepsearch_glm-0.21.1-cp312-cp312-macosx_13_0_x86_64.whl", hash = "sha256:802a59a8a3bea1801bce848d58d19fcdbbcea27d9e2c23f163419d13cdec2345"},
{file = "deepsearch_glm-0.21.1-cp312-cp312-macosx_14_0_arm64.whl", hash = "sha256:1ead7958bc044000a8d43cce53c9b82be0d341b0ca5cf7b39a0c09f9c4fd8ceb"},
{file = "deepsearch_glm-0.21.1-cp312-cp312-macosx_14_0_x86_64.whl", hash = "sha256:312cf2b0b6560c8dfe5331a5a80a0ed5cb409d29ee6cc999a81696774d50f5e7"},
{file = "deepsearch_glm-0.21.1-cp39-cp39-macosx_13_0_arm64.whl", hash = "sha256:4db0a700c08ff2d6285461dc5f4a68ccd36876a59b62131f847dc4be76a85989"},
{file = "deepsearch_glm-0.21.1-cp39-cp39-macosx_13_0_x86_64.whl", hash = "sha256:f1041c44d1a4d1a43a324781795b03edfdfd8076c49a610c4dd384c86f2a6236"},
{file = "deepsearch_glm-0.21.1-cp39-cp39-macosx_14_0_arm64.whl", hash = "sha256:efb0e9678fe07640bd9b6dc07651eaf1f8e5d5602e379b4cf78dbcddc62b50e9"},
{file = "deepsearch_glm-0.21.1-cp39-cp39-macosx_14_0_x86_64.whl", hash = "sha256:f8d46922d74339ec7fd7a6933220ebc36b2ff39738ad9bb74ea55a198dd31b2f"},
]
[package.dependencies]
@ -7228,4 +7220,4 @@ examples = ["langchain-huggingface", "langchain-milvus", "langchain-text-splitte
[metadata]
lock-version = "2.0"
python-versions = "^3.10"
content-hash = "f50b5f6158b688cc25f80253e3cec8e60d852d66a90fe8eb96798ea3c2372019"
content-hash = "036e714aa18f8714c6305cc4de5f6ee9683af27978eb52ad803e848bc490bdc5"

2
pyproject.toml
View File

@ -25,7 +25,7 @@ python = "^3.10"
pydantic = "^2.0.0"
docling-core = "^1.3.0"
docling-ibm-models = "^1.1.7"
deepsearch-glm = "^0.21.0"
deepsearch-glm = "^0.21.1"
filetype = "^1.2.0"
pypdfium2 = "^4.30.0"
pydantic-settings = "^2.3.0"

2
tests/data/2203.01017v2.json
View File

File diff suppressed because one or more lines are too long

2
tests/data/2206.01062.json
View File

File diff suppressed because one or more lines are too long

2
tests/data/2305.03393v1-pg9.json
View File

File diff suppressed because one or more lines are too long

2
tests/data/2305.03393v1-pg9.pages.json
View File

File diff suppressed because one or more lines are too long

2
tests/data/2305.03393v1.json
View File

File diff suppressed because one or more lines are too long

2
tests/data/redp5110.json
View File

File diff suppressed because one or more lines are too long

2
tests/data/redp5695.json
View File

File diff suppressed because one or more lines are too long

14
tests/verify_utils.py
View File

@ -96,10 +96,18 @@ def verify_tables(doc_pred: DsDocument, doc_true: DsDocument):
for i, row in enumerate(true_item.data):
for j, col in enumerate(true_item.data[i]):
print("true: ", true_item.data[i][j])
print("pred: ", pred_item.data[i][j])
assert (
true_item.data[i][j].text == pred_item.data[i][j].text
), "table-cell does not have the same text"
assert (
true_item.data[i][j].type == pred_item.data[i][j].type
), "table-cell does not have the same type"
return True
@ -155,9 +163,9 @@ def verify_conversion_result(
doc_pred_pages, doc_true_pages
), f"Mismatch in PDF cell prediction for {input_path}"
# assert verify_output(
# doc_pred, doc_true
# ), f"Mismatch in JSON prediction for {input_path}"
assert verify_output(
doc_pred, doc_true
), f"Mismatch in JSON prediction for {input_path}"
assert verify_md(
doc_pred_md, doc_true_md