Disable postgres index updation temporarily and update the indexes manually later for insert statement performance – Postgresql

i have about 1300 CSV files with almost 40k of rows in each file, i have written a python code to read the file and convert all the 40k entries into single insert statement to insert in Postgres database.

The psudocode is following

for file in tqdm(files, desc="Processing files"):
    rows = ReadFile(file) # read all 40k rows from the file
    q = GenerateQuery(rows) # convert all rows into single bulk insert statement
    InsertData(q) # Execute the generated query to insert data

The code is fine but there are performance issues, when I start the code with empty table it takes around 2 to 3it/s, but after 15 to 20 files it takes 10 to 12it/s, and then the performance drops exponentially with each 10 to 15 files processing, the per iteration time keeps on increasing even it reaches 40 to 50s/it, it’s hilarious, according to my understanding I have developed the following hypothesis

Since in start table is empty its very easy to update table indexes, so it takes no time for 40k bulk insert records to update indexes but with growing records it become harder and even harder to update indexes in the table with 10m+ records.

My Question is can I temporarily disable index updation of the table, so that after complete data dump I will then manually update the indexes by calling some query in postgres which for now I don’t know if it really exists.