crf | keras implementation of conditional random field | Machine Learning library

I am new to FFmpeg and I'm just getting started. I have the two following commands to take a .mov upload and downsize it and save it out as an .mp4. I also produce a thumbnail version of the same.

What I am doing here seems to work well and I manage to get the .mov file from 12.5mb to about a 1mb .mp4.

My question from here is, can I get away with only providing a .mp4 for the source video, or do I need to provide additional .ogg and/or .webm files too? An if so, how is that best achieved? Would I have to loop that FFmpeg command again to output new .ogg and .webm files?

exec("ffmpeg -i src.mov -ss 00 -to 60 -vf scale=700:-1 -crf 30 -an -movflags +faststart output.mp4");

exec("ffmpeg -i src.mov -ss 00 -to 10 -vf 'scale=128:128:force_original_aspect_ratio=increase, crop=128:128' -crf 30 -an tn_output.mp4");

I am testing a web app and the test runs reliably in headed mode (cypress open) but has errors in headless mode (cypress run), so it's likely a race condition that I cannot resolve. The error message is:

[36819:0223/163815.745047:ERROR:system_services.cc(34)] SetApplicationIsDaemon: Error Domain=NSOSStatusErrorDomain Code=-50 "paramErr: error in user parameter list" (-50)

This error is mentioned again when Cypress creates a video of the incident:

What's the correct ffmpeg setting to prevent the video from 'lazy' loading the video's total duration?

I'm trying to transcode a video using ffmpeg (in NodeJS). When transcoding is finished ffmpeg uploads the video automatically to a bucket (GCP).

Hoverever, the video that ffmpeg produces doesn't have a clear total duration when played in a HTML video player. It's only after x (milli)seconds that the total video duration is known and displayed.

Using the fluent-ffmpeg NodeJS library:

The command below blends input2 into input1 using a binary of *if expressions featuring the between statement's t variable. As such, output = input1, but with input2 at 00:00:03- 00:00:05:

ffmpeg -y -i input1.mkv -i input2.mkv -filter_complex "[1:v][0:v]scale2ref[v1][v0];[v0][v1]blend=all_expr='A*if(between(T, 3, 5), 0, 1)+B*if(between(T, 3, 5), 1, 0)'" -vcodec libx264 -pix_fmt yuv444p -crf 18 -acodec copy output.mkv

My question is how these can be chained together in the case of a set of evaluated conditions. I've attempted the + operator, below, but input1 is no longer the same. It seems distorted by multiple layers of the video.

ffmpeg -y -i input1.mkv -i input2.mkv -filter_complex "[1:v][0:v]scale2ref[v1][v0];[v0][v1]blend=all_expr='A*if(between(T, 3, 5), 0, 1)+B*if(between(T, 3, 5), 1, 0)'+'A*if(between(T, 7, 9), 0, 1)+B*if(between(T, 7, 9), 1, 0)'" -vcodec libx264 -pix_fmt yuv444p -crf 18 -acodec copy output.mkv

How can these be chained together without producing visual artifacts?

This question is a follow-up to Rotem's solution to my earlier question.

I am receiving json with dynamic one node based on sector like "DEL-BOM", "NYC-BOM". trying to parse but body getting null. I just add Dictionary and key as node and class as value but still not getting values in DeserializeObject.

I converted an avi file to a mp4 file with the following command, but the converted mp4 file produced no audio when played with QuickTime (no such problem with other players). I was able to convert mkv to mp4 with the same command without the audio problem.

• In the following json:

I have built a Sentence Boundary Detection Classifier. For the sequence labeling I used a conditional random field. For the hyperparameter optimization I would like to use RandomizedSearchCV. My training data consists of 6 annotated texts. I merge all 6 texts to a tokenlist. For the implementation I followed an example from the documentation. Here my simplified code:

The OBS's Recording settings in Output are:

1. Type: Standard
2. Recording Format: mkv
3. Encoder: NVIDIA NVENC H.264(new)
4. Rate Control: Lossless
5. Keyframe Interval: 0
6. Preset: Max Quality
7. Profile: high [Look-ahead and Psycho Visual Tuning are checked]
8. GPU: 0
9. Max B-Frame: 4

The OBS's Video settings in Advanced are:

1. Renderer: Direct3D 11
2. Color Format: I444
3. Color Space: 709
4. Color Range: Full

Using the settings, OBS can record Windows Desktop in very high quality. However, the MKV file cannot be played by "Movie & TV" app or DivX Player.

OBS will remux the MKV file to MP4, the MP4 file cannot be played by "Movie & TV" app or DivX Pro either. But the MP4 file can be played by Chrome or Edge browser. MKV and MP4 are just containers to store video data, so I think the real problem should be in video data.

How can I record lossless video of Windows Desktop with OBS?

P.S. I have tried to use x264 with CRF 0 to record lossless video, but get the same result.