Bowlus TG-12
The Bowlus TG-12 was a proposed American training glider conceived by Bowlus in the early 1940s.
| TG-12 | |
|---|---|
| Role | Glider |
| National origin | United States |
| Manufacturer | Bowlus Sailplanes |
| Number built | 0 |
| Developed from | Bowlus XBM-5 |
Design and development
The TG-12 was designed as a two-seat training glider with a side-by-side cockpit configuration, based on the company's XBM-5 (company designation). Three TG-12s were ordered on 28 April 1942 with the serials 42-96830/96832 along with a static test airframe, but the order was cancelled on 5 August 1943 without any aircraft being built.[1]
Variants
- Bowlus XTG-12-BS
- 2 completed (Baugher lists only 42-57200), 3 ordered
- Bowlus XTG-12A-BS
- conflicting info., poss. side-by-side one ordered/poss. civil impress.
- Bowlus TG-12A-BS
- [Project] production model, contract cancelled
(Note: the TG-12B was a separate design: a Bowlus built Mitchell M1PU3
Specifications
Data from [1]
General characteristics
- Crew: one
- Length: 24 ft 10 in (7.56 m)
- Wingspan: 45 ft 5 in (13.84 m)
Performance
- Maximum speed: 100 mph (160 km/h, 87 kn)
gollark: If you want to do text you could just use a normal text-only model.
gollark: They generally still require attribution.
gollark: Update update: unfortunately, I cannot achieve low enough validation error to make this actually usable. Probably it would work better if the OCR thing were more accurate (there are issues with spacing), and if I rated memes from a dataset as "good" or "bad" instead of having "good" and "bad" sets from separate places (but this would take too long). I might put the mostly nonfunctional thing on github or something.
gollark: Update on the automatic meme classification thing: after far too much time dealing with various dependencyish issues, my stuff is being run through CLIP and extremely janky OCR then a sentence embedding model. I will begin work on actually implementing a classifier once the script finishes running on everything.
gollark: I mean, it's probably a better metric than picking randomly.
External links
This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.