def per_frame(basedir, func, config):
config_file = os.path.join(basedir, "config.pickle")
cf = pickle.load(open(config_file))
env = util.Environmentz(cf['field_dim_m'], cf['frame_dim_pix'])
FRAMEN = cf['end_f'] - cf['start_f'] + 1
d = np.zeros(FRAMES_TO_ANALYZE, dtype=DTYPE_POS_CONF)
FRAMES_AT_A_TIME = 10
frames = np.arange(FRAMES_TO_ANALYZE)
for frame_subset in util.chunk(frames, FRAMES_AT_A_TIME):
fs = organizedata.get_frames(basedir, frame_subset)
for fi, frame_no in enumerate(frame_subset):
real_x, real_y, conf = func(fs[fi], env, **config)
d[frame_no]['x'] = real_x
d[frame_no]['y'] = real_y
d[frame_no]['confidence'] = conf
return d
def per_frame(basedir, func, config):
config_file = os.path.join(basedir, "config.pickle")
cf = pickle.load(open(config_file))
env = util.Environmentz(cf['field_dim_m'], cf['frame_dim_pix'])
FRAMEN = cf['end_f'] - cf['start_f'] + 1
d = np.zeros(FRAMES_TO_ANALYZE, dtype=DTYPE_POS_CONF)
FRAMES_AT_A_TIME = 10
frames = np.arange(FRAMES_TO_ANALYZE)
for frame_subset in util.chunk(frames, FRAMES_AT_A_TIME):
fs = organizedata.get_frames(basedir, frame_subset)
for fi, frame_no in enumerate(frame_subset):
real_x, real_y, conf = func(fs[fi], env, **config)
d[frame_no]['x'] = real_x
d[frame_no]['y'] = real_y
d[frame_no]['confidence'] = conf
return d
positions_cleaned = positions.copy()
positions_cleaned[invalid_sep] = ((np.nan, np.nan), (np.nan, np.nan),
np.nan, np.nan)
# just take the sanitized data, don't bother interpolating.
# ignore all other positions
valid_frames = np.argwhere(np.isfinite(positions_cleaned['x']))[:-3, 0]
# the -3 above is to deal with some strange offset issues we have where len(positions) != total-number-of-frames
PIX_REGION = 24 # num pixels on either side
ledimgs = np.zeros(
(len(valid_frames), 2, PIX_REGION * 2 + 1, PIX_REGION * 2 + 1),
dtype=np.uint8)
framepos = 0
for frame_chunk in util.chunk(valid_frames, 100):
frames = organizedata.get_frames(basedir, frame_chunk)
for frame_idx, frame in zip(frame_chunk, frames):
for led, field in [(0, 'led_front'), (1, 'led_back')]:
real_pos = positions_cleaned[field][frame_idx]
x, y = env.gc.real_to_image(real_pos[0], real_pos[1])
ledimgs[framepos, led, :, :] = util.extract_region_safe(
frame, int(y), int(x), PIX_REGION, 0)
framepos += 1
ledimgs_mean = np.mean(ledimgs.astype(np.float32), axis=0)
subsamp = 60
led_params_dict = {
positions_cleaned = positions.copy()
positions_cleaned[invalid_sep] = ((np.nan, np.nan),
(np.nan, np.nan), np.nan, np.nan)
# just take the sanitized data, don't bother interpolating.
# ignore all other positions
valid_frames = np.argwhere(np.isfinite(positions_cleaned['x']))[:-3, 0]
# the -3 above is to deal with some strange offset issues we have where len(positions) != total-number-of-frames
PIX_REGION = 24 # num pixels on either side
ledimgs = np.zeros((len(valid_frames), 2, PIX_REGION*2+1,
PIX_REGION*2+1), dtype = np.uint8)
framepos = 0
for frame_chunk in util.chunk(valid_frames, 100):
frames = organizedata.get_frames(basedir, frame_chunk)
for frame_idx, frame in zip(frame_chunk, frames):
for led, field in [(0, 'led_front'),
(1, 'led_back')]:
real_pos = positions_cleaned[field][frame_idx]
x, y = env.gc.real_to_image(real_pos[0], real_pos[1])
ledimgs[framepos, led, :, :] = util.extract_region_safe(frame, int(y), int(x), PIX_REGION, 0)
framepos +=1
ledimgs_mean = np.mean(ledimgs.astype(np.float32),
axis=0)
subsamp = 60
