# Reading tfrecord files

**URL:** <https://discourse.julialang.org/t/reading-tfrecord-files/26316>\
**Category:** Machine Learning\
**Tags:** question\
**Created:** [July 13, 2019, 9:21am UTC](https://discourse.julialang.org/t/reading-tfrecord-files/26316 "2019-07-13T09:21:06Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![jbrea](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/jbrea/32/3879_2.png) [@jbrea](https://discourse.julialang.org/u/jbrea)\
**Post date:** [July 13, 2019, 9:21am UTC](https://discourse.julialang.org/t/reading-tfrecord-files/26316/1 "2019-07-13T09:21:06Z")

</div>

I want to play a bit with the [Youtube 8M dataset](https://research.google.com/youtube8m/download.html). Did somebody already work on reading tfrecord files?

I tried

```julia
julia> using TensorFlow
julia> it = TensorFlow.io.TFRecord.RecordIterator("train0111.tfrecord")
julia> first(it)
351069-element Array{UInt8,1}:
 0x0a
 0x23
 0x0a
 0x0e
 0x0a
 0x02
 ...

```

which I still have to parse with the right proto, I guess. Are there already some tools or generic approaches available for this?

---

<div class="post-metadata">

**Author:** ![findmyway](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/findmyway/32/4946_2.png) [@findmyway](https://discourse.julialang.org/u/findmyway)\
**Post date:** [October 15, 2020, 3:57pm UTC](https://discourse.julialang.org/t/reading-tfrecord-files/26316/2 "2020-10-15T15:57:43Z")

</div>

I made one here: [https://github.com/JuliaReinforcementLearning/TFRecord.jl](https://github.com/JuliaReinforcementLearning/TFRecord.jl) It’s much easier than I thought. I really wish I had made it a year ago! 😅