github/gpt4all
1
0
Fork 0
mirror of https://github.com/nomic-ai/gpt4all.git synced 2025-07-04 02:58:04 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix: embeddings instead of logits!!!

Browse Source
This commit is contained in:
Zach 2023-04-08 17:05:40 +00:00
parent 4b51e6ef37
commit 1c6d2d9622
1 changed files with 5 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
inference.py
View File

@ -115,7 +115,6 @@ def inference(config):
train_outputs["embeddings"] = np.concatenate(train_outputs["embeddings"]) train_outputs["embeddings"] = np.concatenate(train_outputs["embeddings"])
df_train = Dataset.from_dict(train_outputs) df_train = Dataset.from_dict(train_outputs)
df_train = df_train.sort("index")
curr_idx = df_train["index"] curr_idx = df_train["index"]
# compute mask in pyarrow since it's super fast # compute mask in pyarrow since it's super fast
@ -136,11 +135,11 @@ def inference(config):
for batch in tqdm(val_dataloader, disable=local_rank != 0): for batch in tqdm(val_dataloader, disable=local_rank != 0):
batch["input_ids"] = batch["input_ids"].to(f"cuda:{local_rank}") batch["input_ids"] = batch["input_ids"].to(f"cuda:{local_rank}")
batch["labels"] = batch["labels"].to(f"cuda:{local_rank}") batch["labels"] = batch["labels"].to(f"cuda:{local_rank}")
outputs = model(input_ids=batch["input_ids"], labels=batch["labels"]) outputs = model(input_ids=batch["input_ids"], labels=batch["labels"], output_hidden_states=True)
loss = calc_cross_entropy_no_reduction(outputs.logits, batch["labels"]) loss = calc_cross_entropy_no_reduction(outputs.logits, batch["labels"])
val_outputs["loss"].extend(loss) val_outputs["loss"].extend(loss)
logits = outputs.logits embeddings = outputs.hidden_states[-1]
batch_size = batch["input_ids"].shape[0] batch_size = batch["input_ids"].shape[0]
sequence_lengths = [] sequence_lengths = []
# since we use mutiturn with multiple <|endoftext|>, we need to find the place where # since we use mutiturn with multiple <|endoftext|>, we need to find the place where
@ -149,17 +148,17 @@ def inference(config):
indices = torch.where(item == tokenizer.pad_token_id)[0] indices = torch.where(item == tokenizer.pad_token_id)[0]
found = False found = False
for index in indices: for index in indices:
# case where sequence is less than max length
if torch.all(item[index:] == tokenizer.pad_token_id): if torch.all(item[index:] == tokenizer.pad_token_id):
sequence_lengths.append(index) sequence_lengths.append(index)
found = True found = True
break break
# case where sequence is >= max length
# no match found
if not found: if not found:
sequence_lengths.append(len(item) - 1) sequence_lengths.append(len(item) - 1)
sequence_lengths = torch.tensor(sequence_lengths) sequence_lengths = torch.tensor(sequence_lengths)
pooled_logits = logits[torch.arange(batch_size, device=logits.device), sequence_lengths] pooled_logits = embeddings[torch.arange(batch_size, device=embeddings.device), sequence_lengths]
val_outputs["embeddings"].append(pooled_logits) val_outputs["embeddings"].append(pooled_logits)
val_outputs["index"].extend(batch["index"].to(model.device)) val_outputs["index"].extend(batch["index"].to(model.device))
@ -172,7 +171,6 @@ def inference(config):
val_outputs["embeddings"] = np.concatenate(val_outputs["embeddings"]) val_outputs["embeddings"] = np.concatenate(val_outputs["embeddings"])
df_val = Dataset.from_dict(val_outputs) df_val = Dataset.from_dict(val_outputs)
df_val = df_val.sort("index")
curr_idx = df_val["index"] curr_idx = df_val["index"]
# compute mask in pyarrow since it's super fast # compute mask in pyarrow since it's super fast
@ -182,7 +180,6 @@ def inference(config):
filtered_table = table.filter(mask) filtered_table = table.filter(mask)
# convert from pyarrow to Dataset # convert from pyarrow to Dataset
filtered_val = Dataset.from_dict(filtered_table.to_pydict()) filtered_val = Dataset.from_dict(filtered_table.to_pydict())
filtered_val = filtered_val.add_column("embeddings", df_val["embeddings"]) filtered_val = filtered_val.add_column("embeddings", df_val["embeddings"])
filtered_val = filtered_val.add_column("loss", df_val["loss"]) filtered_val = filtered_val.add_column("loss", df_val["loss"])
filtered_val = filtered_val.add_column("is_train", [False] * len(filtered_val)) filtered_val = filtered_val.add_column("is_train", [False] * len(filtered_val))