def write_videos(self, observation, step=None, n=4):
video = torch.clamp(observation[:, :n] + 0.5, 0., 1.).transpose(1, 0)
writer: SummaryWriter = logger.get_tf_summary_writer()
writer.add_video(tag='videos/ground_truth',
vid_tensor=video,
global_step=step,
fps=20)
def log_hparams(params):
writer = logger.get_tf_summary_writer()
params = flatten_dict(params)
filtered_params = dict()
for key, value in params.items():
if type(value) in [int, float, str, bool]: # , torch.Tensor]:
filtered_params[key] = value
hparam_dict = filtered_params
metric_dict = {'Return/Average': float('nan')}
exp, ssi, sei = hparams(hparam_dict, metric_dict)
writer.file_writer.add_summary(exp)
writer.file_writer.add_summary(ssi)
writer.file_writer.add_summary(sei)
for k, v in metric_dict.items():
writer.add_scalar(k, v)
def video_summary(tag, video, step=None, fps=0.5):
writer: SummaryWriter = logger.get_tf_summary_writer()
writer.add_video(tag=tag, vid_tensor=video, global_step=step, fps=fps)