Pagination, bug fixes

• Add a setting to enable output pagination
• Enable convert.py to use mps (but less memory efficient than cpu/cuda)
• Fix bug with inference ram setting
• Fix bug with pdf names with dots in them
• Fix bug with images at the end of blocks

Fix convert.py bug

Fix model device check.

Specify page range

• Make it more clear MPS can't be used with convert.py
• Specify page range in convert with start_page and max_pages

Python 3.12 compatibility

• Remove ray to enable python 3.12 compatibility
• Removing ray frees a lot of VRAM (since we can use torch shared tensors), so on average with convert.py each process takes 3GB VRAM. This enables much higher throughput (was between 4.5GB and 5GB before).

OCR speedups

• Pull in new surya and pdftext versions for speedups in OCR and text extraction, respectively
• Refine heuristics to reduce OCR false positives (and true positives, unfortunately)
• Enable float batch multipliers

Speed improvements

• Enable parallel text extraction, with worker count settings
• Bump surya version to pull in layout/line segmentation speed improvements, and OCR bug fix

Faster OCR

• OCR is now ~2.5x faster, due to improvements in surya

Speed up inference

• (from surya) faster ocr, line detection, layout inference
• Unpin transformers version after testing

Should be significantly faster now, but haven't fully benchmarked, since I'm running low on time this week!

Fix memory leak

• Fix a memory leak (fixed in surya, bumped the version). This caused high CPU memory usage on long docs.
• Improve load_all_models to take device and dtype

Marker v2

Basically a full rewrite!

Main features:

• Extracts and saves images
• Improved table formatting
• Better markdown wrapping
• Better reading order on complex docs
• Improved OCR engine with more language options
• Simple pip package install (no more required system dependencies), so can be used easily on Windows
• Can be used commercially (pymupdf and layoutlmv3 dependencies removed)

It takes ~2x as long to run now, but seems like a decent tradeoff.

See the README for details.