Add 323 files (ocr-tech-mps-integration/tools/prepare_native_spotcheck_bundle.py ...)
0cfd653 verified Download logs/eval_v4_last.log from prosperious/ocr-projects-archive: direct link, hf CLI and curl.
- Browser
- Download file 4.61 kB
-
https://huggingface.co/prosperious/ocr-projects-archive/resolve/main/logs/eval_v4_last.log
- Command line
-
hf download hf://prosperious/ocr-projects-archive/logs/eval_v4_last.log
-
curl -L -o eval_v4_last.log https://huggingface.co/prosperious/ocr-projects-archive/resolve/main/logs/eval_v4_last.log
4.61 kB
| Loading weights: 0%| | 0/1035 [00:00<?, ?it/s] Loading weights: 25%|βββ | 258/1035 [00:00<00:00, 2571.41it/s] Loading weights: 50%|βββββ | 516/1035 [00:00<00:00, 1842.20it/s] Loading weights: 69%|βββββββ | 713/1035 [00:00<00:00, 1878.64it/s] Loading weights: 88%|βββββββββ | 909/1035 [00:00<00:00, 1498.70it/s] Loading weights: 100%|ββββββββββ| 1035/1035 [00:00<00:00, 1683.07it/s] | |
| /home/ampixa/ocr/ocr-tech/tools/deva_rec_torch/export_onnx.py:38: DeprecationWarning: You are using the legacy TorchScript-based ONNX export. Starting in PyTorch 2.9, the new torch.export-based ONNX exporter has become the default. Learn more about the new export logic: https://docs.pytorch.org/docs/stable/onnx_export.html. For exporting control flow: https://pytorch.org/tutorials/beginner/onnx/export_control_flow_model_to_onnx_tutorial.html | |
| torch.onnx.export( | |
| /home/ampixa/ocr/.venv-torch/lib/python3.11/site-packages/transformers/integrations/sdpa_attention.py:124: TracerWarning: Converting a tensor to a Python boolean might cause the trace to be incorrect. We can't record the data flow of Python values, so this value will be treated as a constant in the future. This means that the trace might not generalize to other inputs! | |
| is_causal = q_length > 1 and attention_mask is None and is_causal | |
| batch 1 width 320: T=40 max|diff|=5.96e-07 argmax agree=1.0000 | |
| batch 4 width 480: T=60 max|diff|=1.10e-06 argmax agree=1.0000 | |
| batch 2 width 1200: T=150 max|diff|=6.56e-07 argmax agree=1.0000 | |
| exported and verified: /home/ampixa/ocr/ocr-deva-ft/export/v4-last-onnx | |
| e1v4_chunks: 3000 predictions | |
| e7_kra: 530 predictions | |
| e1b_lines_kfix: 1500 predictions | |
| e2_newar: 1139 predictions | |
| e3_cit_ne: 800 predictions | |
| e3_office_ne: 300 predictions | |
| e3_cit_en: 300 predictions | |
| e3_realtext: 400 predictions | |
| e4_english: 120 predictions | |
| e4_mixed: 120 predictions | |
| e5_handwriting_probe: 450 predictions | |
| e6_nvidia_en: 419 predictions | |
| h1_nepali_cc0_test: 871 predictions | |
| h2_iiit_test: 2000 predictions | |
| h3_indicscan: 835 predictions | |
| h4_dhcd_digits_test: 800 predictions | |
| h4_dhcd_digits_test: 800 predictions | |
| /home/ampixa/ocr/.venv-serve/lib/python3.11/site-packages/paddle/utils/cpp_extension/extension_utils.py:712: UserWarning: No ccache found. Please be aware that recompiling all source files may be required. You can download and install ccache from: https://github.com/ccache/ccache/blob/master/doc/INSTALL.md | |
| warnings.warn(warning_message) | |
| h4_dhcd_digits_test: 800 predictions | |
| set official v3 v4last | |
| e1v4_chunks cp 0.0513 gr 0.0817 ex 0.47 cp 0.0057 gr 0.0098 ex 0.93 cp 0.0063 gr 0.0111 ex 0.93 n=3000 | |
| e7_kra cp 0.0552 gr 0.0996 ex 0.33 cp 0.0086 gr 0.0150 ex 0.89 cp 0.0087 gr 0.0149 ex 0.88 n=530 | |
| e1b_lines_kfix cp 0.0769 gr 0.1265 ex 0.13 cp 0.0257 gr 0.0500 ex 0.54 cp 0.0266 gr 0.0518 ex 0.53 n=1500 | |
| e2_newar cp 0.0911 gr 0.1742 ex 0.06 cp 0.0354 gr 0.0779 ex 0.35 cp 0.0367 gr 0.0802 ex 0.34 n=1139 | |
| e3_cit_ne cp 0.0678 gr 0.1155 ex 0.56 cp 0.0075 gr 0.0125 ex 0.94 cp 0.0083 gr 0.0152 ex 0.92 n=800 | |
| e3_office_ne cp 0.0777 gr 0.1089 ex 0.54 cp 0.0092 gr 0.0138 ex 0.92 cp 0.0111 gr 0.0167 ex 0.90 n=300 | |
| e3_cit_en cp 0.0194 gr 0.0194 ex 0.83 cp 0.0064 gr 0.0064 ex 0.95 cp 0.0055 gr 0.0055 ex 0.96 n=300 | |
| e3_realtext cp 0.0857 gr 0.1388 ex 0.47 cp 0.0120 gr 0.0216 ex 0.89 cp 0.0166 gr 0.0344 ex 0.84 n=400 | |
| e4_english cp 0.0257 gr 0.0257 ex 0.62 cp 0.0164 gr 0.0164 ex 0.72 cp 0.0235 gr 0.0235 ex 0.66 n=120 | |
| e4_mixed cp 0.0678 gr 0.0785 ex 0.36 cp 0.0267 gr 0.0346 ex 0.70 cp 0.0325 gr 0.0415 ex 0.62 n=120 | |
| e5_handwriting_probe cp 0.6465 gr 0.7112 ex 0.01 cp 0.6102 gr 0.7176 ex 0.00 cp 0.5301 gr 0.6529 ex 0.04 n=450 | |
| e6_nvidia_en cp 0.1062 gr 0.1062 ex 0.71 cp 0.0284 gr 0.0284 ex 0.84 cp 0.0357 gr 0.0357 ex 0.81 n=419 | |
| h1_nepali_cc0_test cp 0.4075 gr 0.5477 ex 0.20 cp 0.6886 gr 0.9664 ex 0.04 cp 0.1554 gr 0.2426 ex 0.55 n=871 | |
| h2_iiit_test cp 0.4944 gr 0.6582 ex 0.07 cp 0.6051 gr 0.8296 ex 0.03 cp 0.0911 gr 0.1521 ex 0.64 n=2000 | |
| h3_indicscan cp 0.2520 gr 0.3649 ex 0.38 cp 0.4580 gr 0.6264 ex 0.14 cp 0.1337 gr 0.2053 ex 0.59 n=835 | |
| h4_dhcd_digits_test cp 0.7185 gr 0.7185 ex 0.02 cp 0.5295 gr 0.5295 ex 0.05 cp 0.0021 gr 0.0021 ex 0.98 n=800 | |
| EVAL_V4_DONE | |