From 2422980acd1abee6b4ab2d146718c6d869ea770f Mon Sep 17 00:00:00 2001
From: liangrz <liangrz15@mails.tsinghua.edu.cn>
Date: Wed, 10 Jun 2020 19:13:06 +0800
Subject: [PATCH] fix the bug of nan gradient
---
convlab2/policy/gdpl/gdpl.py | 3 +++
convlab2/policy/mle/loader.py | 2 +-
convlab2/policy/pg/pg.py | 3 +++
3 files changed, 7 insertions(+), 1 deletion(-)
diff --git a/convlab2/policy/gdpl/gdpl.py b/convlab2/policy/gdpl/gdpl.py
index c40be3b3..7d213be5 100755
--- a/convlab2/policy/gdpl/gdpl.py
+++ b/convlab2/policy/gdpl/gdpl.py
@@ -176,6 +176,9 @@ class GDPL(Policy):
# backprop
surrogate.backward()
+
+ for p in self.policy.parameters():
+ p.grad[p.grad != p.grad] = 0.0
# gradient clipping, for stability
torch.nn.utils.clip_grad_norm(self.policy.parameters(), 10)
# self.lock.acquire() # retain lock to update weights
diff --git a/convlab2/policy/mle/loader.py b/convlab2/policy/mle/loader.py
index b9b62efd..02119d2a 100755
--- a/convlab2/policy/mle/loader.py
+++ b/convlab2/policy/mle/loader.py
@@ -17,7 +17,7 @@ class ActMLEPolicyDataLoader():
data_loader = ActPolicyDataloader(dataset_dataloader=MultiWOZDataloader())
for part in ['train', 'val', 'test']:
self.data[part] = []
- raw_data = data_loader.load_data(data_key=part, role='system')[part]
+ raw_data = data_loader.load_data(data_key=part, role='sys')[part]
for belief_state, context_dialog_act, terminated, dialog_act in \
zip(raw_data['belief_state'], raw_data['context_dialog_act'], raw_data['terminated'], raw_data['dialog_act']):
diff --git a/convlab2/policy/pg/pg.py b/convlab2/policy/pg/pg.py
index b7c6efc1..82df69a8 100755
--- a/convlab2/policy/pg/pg.py
+++ b/convlab2/policy/pg/pg.py
@@ -126,6 +126,9 @@ class PG(Policy):
# backprop
surrogate.backward()
+
+ for p in self.policy.parameters():
+ p.grad[p.grad != p.grad] = 0.0
# gradient clipping, for stability
torch.nn.utils.clip_grad_norm(self.policy.parameters(), 10)
# self.lock.acquire() # retain lock to update weights
--
GitLab