aefc81cb2dfaf8b51b7d917aba743c73078ecad8,rl_loop/train_and_validate.py,,train,#,48

Before Change

    new_model_num = model_num + 1
    new_model_name = shipname.generate(new_model_num)
    print("New model will be {}".format(new_model_name))
    training_file = os.path.join(
        fsdb.golden_chunk_dir(), str(new_model_num) + ".tfrecord.zz")
    while not gfile.Exists(training_file):
        print("Waiting for", training_file)
        time.sleep(1 * 60)
    save_file = os.path.join(fsdb.models_dir(), new_model_name)

After Change

           "--flagfile=rl_loop/distributed_flags",
           "--export_path={}".format(save_file)]

    completed_process = mask_flags.run(cmd)
    if completed_process.returncode > 0:
        print("Training failed!")
        return completed_process

    // Persist the checkpoint two ways:
    // Freeze the .ckpt file in the work_dir for the TPU selfplayers
    // Freeze a non-tpu version of the graph for later GPU use.
    dual_net.freeze_latest_checkpoint_tpu()
    dual_net.freeze_graph(save_file)

    return completed_process

Instances