def main(start_model_path,
start_type,
save_dir,
modeling_option,
target_idx_,
num_gpu=1):
num_gpu = int(num_gpu)
tf_logging.info("train_from : nli_ex")
hp = HPCommon()
nli_setting = BertNLI()
set_level_debug()
reset_root_log_handler()
train_config = NLIPairingTrainConfig()
train_config.num_gpu = num_gpu
tf_logging.info("loading batches")
data = get_nli_data(hp, nli_setting)
def init_fn(sess):
return init_fn_generic(sess, start_type, start_model_path)
class LMSConfig2(LMSConfigI):
num_tags = 3
target_idx = target_idx_
use_embedding_out = True
per_layer_component = 'linear'
train_LMS(hp, train_config, LMSConfig2(), save_dir, data, modeling_option,
init_fn)
def get_params(start_model_path, start_type, info_fn_name, num_gpu):
hp = hyperparams.HPSENLI3()
nli_setting = BertNLI()
set_level_debug()
train_config = NLIExTrainConfig()
train_config.num_gpu = num_gpu
train_config.save_train_payload = True
tokenizer = get_tokenizer()
tf_logging.info("Intializing dataloader")
data_loader = get_modified_data_loader(tokenizer, hp.seq_max,
nli_setting.vocab_filename)
tf_logging.info("loading batches")
data = get_nli_data(hp, nli_setting)
def init_fn(sess):
start_type_generic = {
'nli': 'cls',
'nli_ex': 'cls_ex',
'bert': 'bert',
'cold': 'cold'
}[start_type]
return init_fn_generic(sess, start_type_generic, start_model_path)
informative_fn = get_informative_fn_by_name(info_fn_name)
return data, data_loader, hp, informative_fn, init_fn, train_config
def run(explain_tag, method_name, model_path):
hp = hyperparams.HPSENLI3()
nli_setting = BertNLI()
target_class = enlidef.get_target_class(explain_tag)
data = get_nli_data(hp, nli_setting)
train_batches, dev_batches = data
flat_dev_batches = flatten_filter_batches(dev_batches, target_class)[:2000]
if method_name in ['deletion_seq', "random", 'idf', 'deletion', 'LIME']:
predictor = BaselineExPredictor(hp, nli_setting, model_path,
method_name)
acc_list = eval_fidelity_with_ex_predictor(predictor, flat_dev_batches,
explain_tag)
elif method_name.startswith("nli_ex"):
modeling_option = "co"
predictor = NLIExPredictor(hp, nli_setting, model_path,
modeling_option)
acc_list = eval_fidelity_with_ex_predictor(predictor, flat_dev_batches,
explain_tag)
elif method_name in [
"elrp", "deeplift", "saliency", "grad*input", "intgrad"
]:
acc_list = eval_fidelity_gradient(hp, nli_setting, flat_dev_batches,
explain_tag, method_name, model_path)
else:
raise Exception(
"method_name={} is not in the known method list.".format(
method_name))
print(method_name)
for num_delete in sorted(acc_list.keys()):
print("{}\t{}".format(num_delete, acc_list[num_delete]))
def train_nil_from_bert(model_path, save_dir):
def load_fn(sess, model_path):
return load_model_w_scope(sess, model_path, "bert")
max_steps = 61358
max_steps = 36250
hp = hyperparams.HPSENLI3()
set_level_debug()
nli_setting = BertNLI()
data = get_snli_data(hp, nli_setting)
n_gpu = 2
return train_nli_multi_gpu(hp, nli_setting, save_dir, max_steps, data,
model_path, load_fn, n_gpu)
def main(start_model_path, modeling_option, num_gpu=1):
num_gpu = int(num_gpu)
hp = HPCommon()
nli_setting = BertNLI()
set_level_debug()
reset_root_log_handler()
train_config = NLIPairingTrainConfig()
train_config.num_gpu = num_gpu
def init_fn(sess):
return init_fn_generic(sess, "as_is", start_model_path)
data = get_nli_data(hp, nli_setting)
do_predict(hp, train_config, data, LMSConfig(), modeling_option, init_fn)
def run(args):
hp = hyperparams.HPSENLI3()
nli_setting = BertNLI()
data_loader = get_modified_data_loader2(hp, nli_setting)
predict_nli_ex(
hp,
nli_setting,
data_loader,
args.tag,
args.data_id,
args.model_path,
args.run_name,
args.modeling_option,
)
def main(start_model_path, modeling_option, input_path, save_name, num_gpu=1):
num_gpu = int(num_gpu)
hp = HPCommon()
nli_setting = BertNLI()
set_level_debug()
reset_root_log_handler()
train_config = NLIPairingTrainConfig()
train_config.num_gpu = num_gpu
def init_fn(sess):
return init_fn_generic(sess, "as_is", start_model_path)
batches = get_batches(input_path, nli_setting, HPCommon.batch_size)
output_d = do_predict(hp, train_config, batches, LMSConfig(),
modeling_option, init_fn)
save_to_pickle(output_d, save_name)
def main(start_model_path, start_type, save_dir, modeling_option, num_gpu=1):
num_gpu = int(num_gpu)
tf_logging.info("train_from : nli_ex")
hp = HPCommon()
nli_setting = BertNLI()
set_level_debug()
reset_root_log_handler()
train_config = NLIPairingTrainConfig()
train_config.num_gpu = num_gpu
tf_logging.info("loading batches")
data = get_nli_data(hp, nli_setting)
def init_fn(sess):
return init_fn_generic(sess, start_type, start_model_path)
train_LMS(hp, train_config, LMSConfig(), save_dir, data, modeling_option,
init_fn)
def train_nil_from(model_path, save_dir, resume=False):
print("Load model path : ", model_path)
print("Save dir : ", save_dir)
def load_fn(sess, model_path):
if not resume:
return load_model_w_scope(sess, model_path, "bert")
else:
return load_model(sess, model_path)
steps = 67000
hp = hyperparams.HPSENLI3()
nli_setting = BertNLI()
data = get_nli_data(hp, nli_setting)
set_level_debug()
hp = hyperparams.HPSENLI3()
n_gpu = 2
return train_nli_multi_gpu(hp, nli_setting, save_dir, steps, data,
model_path, load_fn, n_gpu)
def main(start_model_path, start_type, save_dir,
modeling_option, num_gpu=1):
num_gpu = int(num_gpu)
tf_logging.info("Train with MLP")
hp = HPCommon()
hp.per_layer_component = 'mlp'
nli_setting = BertNLI()
set_level_debug()
reset_root_log_handler()
train_config = NLIPairingTrainConfig()
train_config.num_gpu = num_gpu
lms_config = LMSConfig()
tf_logging.info("loading batches")
data = get_nli_data(hp, nli_setting)
def init_fn(sess):
return init_fn_generic(sess, start_type, start_model_path)
train_LMS(hp, train_config, lms_config, save_dir, data, modeling_option, init_fn)
def run(args):
hp = hyperparams.HPSENLI3()
nli_setting = BertNLI()
data_loader = get_modified_data_loader2(hp, nli_setting)
if args.method_name in [
'deletion_seq', "random", 'idf', 'deletion', 'LIME'
]:
predictor = nli_baseline_predict
elif args.method_name in [
"elrp",
"deeplift",
"saliency",
"grad*input",
"intgrad",
]:
predictor = nli_attribution_predict
else:
raise Exception(
"method_name={} is not in the known method list.".format(
args.method_name))
predictor(hp, nli_setting, data_loader, args.tag, args.method_name,
args.data_id, args.sub_range, args.model_path)
def run(args):
hp = hyperparams.HPGenEx()
nli_setting = BertNLI()
if args.method_name in [
'deletion_seq', "random", 'idf', 'deletion', 'LIME',
'term_deletion', 'replace_token', 'term_replace'
]:
predictor = baseline_predict
elif args.method_name in [
"elrp", "deeplift", "saliency", "grad*input", "intgrad"
]:
predictor = nli_attribution_predict
else:
raise Exception(
"method_name={} is not in the known method list.".format(
args.method_name))
save_name = "{}_{}".format(args.data_name, args.method_name)
data = load_as_simple_format(args.data_name)
explains: List[np.array] = predictor(hp, nli_setting, data,
args.method_name, args.model_path)
save_to_pickle(explains, save_name)
def train_nil_from(save_dir, model_path, load_fn, max_steps):
hp = hyperparams.HPSENLI3()
set_level_debug()
nli_setting = BertNLI()
data = get_nli_data(hp, nli_setting)
train_nli(hp, nli_setting, save_dir, max_steps, data, model_path, load_fn)