export model

egs/aishell/ASR/RESULTS.md

@@ -1,5 +1,47 @@
 ## Results
 
+### Aishell training results (Conformer-MMI)
+#### 2021-12-01
+(Pingfeng Luo): Result of <https://github.com/k2-fsa/icefall/pull/123>
+
+The tensorboard log for training is available at <https://tensorboard.dev/experiment/dyp3vWE9RE6SkqBAgLJjUw/>
+
+And pretrained model is available at <https://huggingface.co/pfluo/icefall_aishell_model>
+
+The best decoding results (CER) are listed below, we got this results by averaging models from epoch 20 to 49, and using `attention-decoder` decoder with num_paths equals to 100.
+
+||test|
+|--|--|
+|CER| 5.12% |
+
+||lm_scale|attention_scale|
+|--|--|--|
+|test|1.5|0.5|
+
+You can use the following commands to reproduce our results:
+
+```bash
+git clone https://github.com/k2-fsa/icefall
+cd icefall
+
+cd egs/aishell/ASR
+./prepare.sh
+
+export CUDA_VISIBLE_DEVICES="0,1,2,3,4,5,6,7,8"
+python conformer_ctc/train.py --bucketing-sampler True \
+                              --max-duration 200 \
+			       --start-epoch 0 \
+                              --num-epoch 50 \
+                              --world-size 8
+
+python3 conformer_ctc/decode.py --nbest-scale 0.5 \
+                               --epoch 49 \
+                               --avg 20 \
+                               --method attention-decoder \
+                               --max-duration 20 \
+                               --num-paths 100
+```
+
 ### Aishell training results (Conformer-CTC)
 #### 2021-11-16
 (Wei Kang): Result of https://github.com/k2-fsa/icefall/pull/30

egs/aishell/ASR/conformer_mmi/decode.py

@@ -542,8 +542,14 @@ def main():
             if start >= 0:
                 filenames.append(f"{params.exp_dir}/epoch-{i}.pt")
         logging.info(f"averaging {filenames}")
-        model.to(device)
+        model.load_state_dict(average_checkpoints(filenames))
-        model.load_state_dict(average_checkpoints(filenames, device=device))
+
+    if params.export:
+        logging.info(f"Export averaged model to {params.exp_dir}/pretrained.pt")
+        torch.save(
+            {"model": model.state_dict()}, f"{params.exp_dir}/pretrained.pt"
+        )
+        return
 
     model.to(device)
     model.eval()