Large Tif to parquet

[ChocopieKewpie]

I am quite new to using this, but I am trying to convert a large ~>200gb tif and export it into an xyz in the form of a parquet.
This code works with small tifs, but I seem to run out of memory when writing the parquet file with the large raster.

f = 'somelargetif.tif' #set file 

#converting crs and type to vrt
epsg_to = 4326
with rasterio.open(f) as src:
    with WarpedVRT(src,resampling=1,src_crs=src.crs,crs=crs.CRS.from_epsg(epsg_to)) as vrt:
        raster= rioxarray.open_rasterio(vrt, chunks=True, cache=False).chunk({'y':'auto','x':'auto','band':1})
        raster= raster.squeeze().drop("spatial_ref").drop("band")
        raster.name= "data"
        raster=raster.to_dataset()
raster

pf= 'raster.pqt'

raster.to_dataframe().to_parquet(pf, engine='auto', compression='gzip') #i seem to run out of memory here, perhaps utilize windows?

I've looked at #501 using the isel_window, but I can't wrap head around setting the window in the loop,

BASE_PATH= 'homedir/' #set file directory

da = rioxarray.open_rasterio(f"{BASE_PATH}reproj_dem.tif", cache=False, mask_and_scale=True)

windows = [] #how do you set the window size? 
out_data = []
for window in windows:
    subset = da.rio.isel_window(window)
    out_data.append(
        subset
            .to_dataframe()
            .reset_index()
            .drop_vars("spatial_ref")
            .dropna(how="all", subset=list(subset.data_vars))
   )

pd.concat(out_data).to_parquet(f"{BASE_PATH}tiff_selection.parquet")

et = time.time()
elapsed_time = et - st

Sorry if this question seems basic, but I greatly appreciate all the help.

[snowman2]

See: https://rasterio.readthedocs.io/en/latest/topics/windowed-rw.html

Note: You don't have to use windows. You can also iterate over slices and use isel in array.

[ChocopieKewpie]

Thank you very much for your help :)