Skip to content

Commit

Permalink
Updated README
Browse files Browse the repository at this point in the history
  • Loading branch information
TenzinCHW committed May 15, 2024
1 parent 3635308 commit 6a22115
Showing 1 changed file with 31 additions and 15 deletions.
46 changes: 31 additions & 15 deletions README.md
Original file line number Diff line number Diff line change
@@ -1,15 +1,6 @@
# NeuralMaxEnt

## Data
I have used this data for initial analysis [pvc3 data](https://crcns.org/data-sets/vc/pvc-3/about)

## Experiments and analysis
To run experiments on the above dataset, run the `blanch_expt.jl` file with the following command:
```julia blanche_expt.jl <binsz> <maxiter> <numsplt> <dev> <batchsize>```

For these experiments, `binsz` ranged from 500 to 5000 (in microseconds), `maxiter` was set to 100, `numsplit` was set to 10, `dev` is just the device ID for different runs and `batchsize` was set to 10000.


## DrWatson install instructions
This code base is using the [Julia Language](https://julialang.org/) and
[DrWatson](https://juliadynamics.github.io/DrWatson.jl/stable/)
to make a reproducible scientific project named
Expand All @@ -30,10 +21,35 @@ To (locally) reproduce this project, do the following:
This will install all necessary packages for you to be able to run the scripts and
everything should work out of the box, including correctly finding local paths.

You may notice that most scripts start with the commands:
```julia
using DrWatson
@quickactivate "NeuralMaxEnt"
## Estimating entropy with CDM in octave
First install [octave](https://octave.org/download).
Then, start octave and innstall `statistics`, `struct` and `parallel` with
```
which auto-activate the project and enable local path handling from DrWatson.
pkg install -forge statistics
pkg install -forge struct
pkg install -forge parallel
```

## Data
I have used these two datasets for my experiments: [pvc3](https://crcns.org/data-sets/vc/pvc-3/about)
and [this one from Lamberti](https://datadryad.org/stash/dataset/doi:10.5061/dryad.p5hqbzkqj).

## Experiments and analysis
To run experiments on the above datasets, place their files into a `data/exp_raw` directory and
run the `blanch_expt.jl` file with the following command:
```julia blanche_expt.jl <binsz> <maxiter> <numsplit> <dev> <batchsize>```

### Heatmaps
For these experiments, `binsz` ranged from 500 to 6000 (in microseconds), `maxiter` was set to 100,
`numsplit` was set to 1, `dev` is the device ID and `batchsize` was set to 10000.
If GPU runs out of memory, please lower the `batchsize` variable or use smaller window sizes.

### Entropy vs window size
We chose a fixed `binsz` and varied `winsz` from 1 to some large value (typically up to
a timescale that we're interested in based on the chosen bin size).

## Citations
Blanche, Tim (2009): Multi-neuron recordings in primary visual cortex. CRCNS.org.
http://dx.doi.org/10.6080/K0MW2F2J
Lamberti, M., Hess, M., Dias, I. et al. Maximum entropy models provide functional connectivity estimates in neural networks. Sci Rep 12, 9656 (2022). https://doi.org/10.1038/s41598-022-13674-4

0 comments on commit 6a22115

Please sign in to comment.