Skip to content

Commit

Permalink
Merge pull request #86 from OCR-D/kba-patch-1
Browse files Browse the repository at this point in the history
Update min/med/max table, ht @bertsky, OCR-D/ocrd-website#92
  • Loading branch information
stweil authored Jun 10, 2020
2 parents c580432 + 5eb12bf commit 650eace
Showing 1 changed file with 16 additions and 11 deletions.
27 changes: 16 additions & 11 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -299,12 +299,13 @@ docker pull ocrd/all:maximum
Usage is the same [as if you had built the image yourself](#results).

This table lists which tag contains which module:

| Module | `minimum` | `medium` | `maximum` |
| ----- | ---- | ---- | ---- |
| core | ☑ | ☑ | ☑ |
| ocrd_cis | ☑ | ☑ | ☑ |
| ocrd_fileformat | ☑ | ☑ | ☑ |
| ocrd_im6convert | ☑ | ☑ | ☑ |
| ocrd_pagetopdf | ☑ | ☑ | ☑ |
| ocrd_repair_inconsistencies | ☑ | ☑ | ☑ |
| ocrd_tesserocr | ☑ | ☑ | ☑ |
| tesserocr | ☑ | ☑ | ☑ |
Expand All @@ -318,13 +319,12 @@ This table lists which tag contains which module:
| ocrd_segment | - | ☑ | ☑ |
| tesseract | - | ☑ | ☑ |
| ocrd_anybaseocr | - | - | ☑ |
| ocrd_kraken | - | - | ☑ |
| ocrd_ocropy | - | - | ☑ |
| ocrd_pc_segmentation | - | - | ☑ |
| ocrd_typegroups_classifier | - | - | ☑ |
| sbb_textline_detector | - | - | ☑ |
| cor-asv-fst | - | - | - |
| clstm | - | - | - |
| ocrd_kraken | - | - | - |
| ocrd_ocropy | - | - | - |
| cor-asv-fst | - | - | ☑ |

**Note**: The following modules have been disabled by default and can only be
enabled by explicitly setting `OCRD_MODULES` or `DISABLED_MODULES`:
Expand Down Expand Up @@ -359,16 +359,21 @@ Modules may require mutually exclusive sets of dependent packages.
`pip` does not even stop or resolve conflicts – it merely warns!

- Tensorflow:
* `tensorflow-gpu==1.14.0` (required by ocrd_calamari and ocrd_anybaseocr)
* `tensorflow` (required by cor-asv-ann and ocrd_keraslm)
* version 2 (required by ocrd_anybaseocr and ocrd_pc_segmentation)
* version 1 (required by cor-asv-ann, ocrd_keraslm, ocrd_calamari and sbb_textline_detector)

The temporary solution is to require different package names:
- `tensorflow>=2`
- `tensorflow-gpu==1.15.*`

Both can be installed in parallel in different versions, but may depend on a mutually exclusive set of `tensorboard` and `tensorflow_estimator`.
Both can be installed in parallel in different versions, but may depend on a mutually exclusive set of `tensorboard` and `tensorflow_estimator`. (However, these latter conflicts are only cosmetic, as they are only needed for development.)

Moreover, in the future, some modules (but not others) may depend on `tensorflow>=2.0`, which again is incompatible.
- OpenCV:
* `opencv-python-headlesss` (required by core and others, avoids pulling in X11 libraries)
* `opencv-python` (required by ocrd_anybaseocr)
* custom build on ARM...
* `opencv-python` (probably dragged in by third party packages)

As long as we keep reinstalling the headless variant and no such package attempts GUI, we should be fine.
Custom build (as needed for ARM) under the _module_ `opencv-python` already creates the headless variant.

- ...

Expand Down

0 comments on commit 650eace

Please sign in to comment.