bugfixes and adds a section to convert optim files

openfold/utils/import_weights.py

@@ -14,6 +14,7 @@
 # limitations under the License.
 
 import re
+import logging
 from enum import Enum
 from dataclasses import dataclass
 from functools import partial
@@ -669,6 +670,7 @@ def import_jax_weights_(model, npz_path, version="model_1"):
 
 def convert_deprecated_v1_keys(state_dict):
     """Update older OpenFold model weight names to match the current model code."""
+    logging.warning('converting keys...')
 
     replacements = {
         'template_angle_embedder': 'template_single_embedder',
@@ -686,17 +688,22 @@ def convert_deprecated_v1_keys(state_dict):
     converted_state_dict = {}
     for key, value in state_dict.items():
         # For each match, look-up replacement value in the dictionary
-        new_key = convert_key_re.sub(lambda m: replacements[m.group()], key)
+        new_key = convert_key_re.sub(lambda m: replacements[m.group(1)], key)
+        ### DEBUG: remove before final commit
         if key == 'template_angle_embedder.linear_1.weight':
-            print(f'old key: {key}, new_key: {new_key}')
+            logging.warning(f'old key: {key}, new_key: {new_key}')
+        ### DEBUG: remove before final commit
 
         # Add prefix for template layers 
         template_match = re.match(template_emb_re, new_key)
         if template_match:
             prefix = template_match.group(1)
             new_key = f'{prefix if prefix else ""}template_embedder.{template_match.group(4)}'
+        # DEBUG: remove before final commit
         if key == 'template_angle_embedder.linear_1.weight':
-            print(f'old key: {key}, new_key: {new_key}')
+            breakpoint()
+            logging.warning(f'old key: {key}, new_key: {new_key}')
+        ### DEBUG: remove before final commit
 
         converted_state_dict[new_key] = value
 

scripts/convert_v1_to_v2_weights.py

@@ -15,6 +15,7 @@
 # Converts OpenFold .pt checkpoints into AlphaFold .npz ones, which can then be
 # used to run inference using DeepMind's JAX code.
 
+import logging
 import argparse
 import os
 import shutil
@@ -23,47 +24,61 @@ import torch
 from openfold.utils.import_weights import convert_deprecated_v1_keys
 from zero_to_fp32 import get_optim_files, parse_optim_states, get_model_state_file
 
-def get_latest_checkpoint_dir(checkpoint_dir):
-    # Based on zero_to_fp32.get_fp32_state_dict_from_zero_checkpoint
-    latest_path = os.path.join(checkpoint_dir, 'latest')
-    if os.path.isfile(latest_path):
-        with open(latest_path, 'r') as fd:
-            tag = fd.read().strip()
-    else:
-        raise ValueError(f"Unable to find 'latest' file at {latest_path}")
-
-    return os.path.join(checkpoint_dir, tag)
-
 
 def convert_v1_to_v2_weights(args):
-    # TODO can we use zero_to_fp32.get_fp32_state_dict_from_zero_checkpoint here?
-
     checkpoint_path = args.input_ckpt_path
     is_dir = os.path.isdir(checkpoint_path)
     if is_dir:
         # A DeepSpeed checkpoint
-        ds_checkpoint_path = get_latest_checkpoint_dir(checkpoint_path)
+        logging.info(
+            'Converting checkpoint found at {args.input_checkpoint_path}')
         state_dict_key = 'module'
-        optim_files = get_optim_files(ds_checkpoint_path)
-        zero_stage, _, _ = parse_optim_states(optim_files, ds_checkpoint_path)
-        model_file = get_model_state_file(ds_checkpoint_path, zero_stage)
+        latest_path = os.path.join(checkpoint_path, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+        ds_checkpoint_dir = os.path.join(checkpoint_path, tag)
+        model_output_path = os.path.join(args.output_ckpt_path, tag)
+        optim_files = get_optim_files(ds_checkpoint_dir)
+        zero_stage, _, _ = parse_optim_states(optim_files, ds_checkpoint_dir)
+        model_file = get_model_state_file(ds_checkpoint_dir, zero_stage)
     else:
         # A Pytorch Lightning checkpoint
         state_dict_key = 'state_dict'
+        model_output_path = args.output_ckpt_path
         model_file = checkpoint_path
 
     model_dict = torch.load(model_file, map_location=torch.device('cpu'))
-    model_dict[state_dict_key] = convert_deprecated_v1_keys(model_dict[state_dict_key])
+    model_dict[state_dict_key] = convert_deprecated_v1_keys(
+        model_dict[state_dict_key])
 
     if 'ema' in model_dict:
         ema_state_dict = model_dict['ema']['params']
-        model_dict['ema']['params'] = convert_deprecated_v1_keys(ema_state_dict)
+        model_dict['ema']['params'] = convert_deprecated_v1_keys(
+            ema_state_dict)
 
     if is_dir:
+        param_shapes = convert_deprecated_v1_keys(
+            model_dict['param_shapes'][0])
+        model_dict['param_shapes'] = [param_shapes]
+
         shutil.copytree(checkpoint_path, args.output_ckpt_path)
-        out_fname = os.path.join(args.output_ckpt_path, os.path.basename(model_file))
+        out_fname = os.path.join(
+            model_output_path, os.path.basename(model_file))
+
+        for optim_file in optim_files:
+            optim_dict = torch.load(optim_file)
+            new_optim_dict = optim_dict.copy()
+            new_optim_dict['optimizer_state_dict']['param_slice_mappings'][0] = convert_deprecated_v1_keys(
+                optim_dict['optimizer_state_dict']['param_slice_mappings'][0])
+            out_optim_fname = os.path.join(
+                model_output_path, os.path.basename(optim_file))
+            torch.save(new_optim_dict, out_optim_fname)
     else:
-        out_fname = args.output_ckpt_path
+        out_fname = model_output_path
 
     torch.save(model_dict, out_fname)
 
@@ -75,4 +90,4 @@ if __name__ == "__main__":
 
     args = parser.parse_args()
 
-    convert_v1_to_v2_weights(args)
\ No newline at end of file
+    convert_v1_to_v2_weights(args)

train_openfold.py

@@ -39,6 +39,7 @@ from scripts.zero_to_fp32 import (
     get_fp32_state_dict_from_zero_checkpoint,
     get_global_step_from_zero_checkpoint
 )
+from scripts.zero_to_fp32 import get_optim_files, parse_optim_states, get_model_state_file
 
 from openfold.utils.logger import PerformanceLoggingCallback
 
@@ -288,6 +289,16 @@ def main(args):
             sd = torch.load(args.resume_from_ckpt)
             last_global_step = int(sd['global_step'])
         model_module.resume_last_lr_step(last_global_step)
+
+        ### DEBUG:
+        ds_checkpoint_dir = os.path.join(args.resume_from_ckpt, 'global_step210')
+        optim_files = get_optim_files(ds_checkpoint_dir)
+        zero_stage, _, _ = parse_optim_states(optim_files, ds_checkpoint_dir)
+        model_file = get_model_state_file(ds_checkpoint_dir, zero_stage)
+
+        model_dict = torch.load(model_file, map_location=torch.device('cpu'))
+        ###
+
         logging.info("Successfully loaded last lr step...")
     if(args.resume_from_ckpt and args.resume_model_weights_only):
         if(os.path.isdir(args.resume_from_ckpt)):