Upload 10 files

LICENSE.txt

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2025 Timur Hromek
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

checkpoints_translation/checkpoint_epoch_1_valloss_4.8841.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2debe09caf3aa76cf341d4985eb301077ee5b790fc4d5bc8348b45b93797316
+size 260563794

checkpoints_translation/checkpoint_epoch_2_valloss_4.3551.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97472d527a0abd2777adea5d272d8c98d2a1db30e7f3de809e639d91fd31b35a
+size 260563794

checkpoints_translation/checkpoint_epoch_3_valloss_4.1226.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9dee8be3fb660152565e108f09ac660afa0defaef8b5c50a43218bad39777cc
+size 260563794

checkpoints_translation/checkpoint_epoch_4_valloss_nan.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf50ebd062d5abf0ce49936d7fbe129c271807b70dc6a9db5a5c305b2f668606
+size 260562255

checkpoints_translation/checkpoint_epoch_5_valloss_nan.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e77b63efc8ba8ca6dc99b25e7ed28e2796bed38f5f67ee5572bc4aa897b9b12
+size 260562255

translate_train.py

@@ -0,0 +1,255 @@
+import os
+import math
+import logging
+from pathlib import Path
+from datetime import datetime
+
+import torch
+import torch.nn as nn
+from torch.nn.utils.rnn import pad_sequence
+from torch.utils.data import DataLoader, Dataset
+
+from datasets import load_dataset
+from tokenizers import Tokenizer
+from tokenizers.models import BPE
+from tokenizers.trainers import BpeTrainer
+from tokenizers.pre_tokenizers import Whitespace
+from tqdm import tqdm
+
+# --- Configuration ---
+CONFIG = {
+    "SRC_LANG": "en",
+    "TGT_LANG": "zh",
+    "TOKENIZER_FILE": "opus_en_zh_tokenizer.json",
+    "MAX_SEQ_LEN": 128,
+    "VOCAB_SIZE": 32000,
+    "DIM": 256,
+    "ENCODER_LAYERS": 4,
+    "DECODER_LAYERS": 4,
+    "N_HEADS": 8,
+    "FF_DIM": 512,
+    "DROPOUT": 0.1,
+    "BATCH_SIZE": 64,
+    "LEARNING_RATE": 5e-4,
+    "NUM_EPOCHS": 5,
+    "CHECKPOINT_DIR": "checkpoints_translation",
+}
+
+logging.basicConfig(level=logging.INFO, format='%(asctime)s - %(levelname)s - %(message)s')
+
+# --- Tokenizer Manager ---
+class TokenizerManager:
+    # ... (No changes needed in this class)
+    def __init__(self, config):
+        self.config = config
+        self.tokenizer_path = Path(self.config["TOKENIZER_FILE"])
+        self.special_tokens = ["<unk>", "<pad>", "<s>", "</s>"]
+    def get_text_iterator(self):
+        dataset = load_dataset(f"Helsinki-NLP/opus-100", f"{self.config['SRC_LANG']}-{self.config['TGT_LANG']}", split="train", streaming=True)
+        for item in dataset: yield item['translation'][self.config['SRC_LANG']]; yield item['translation'][self.config['TGT_LANG']]
+    def train_tokenizer(self):
+        logging.info("Training a new tokenizer...")
+        tokenizer = Tokenizer(BPE(unk_token="<unk>")); tokenizer.pre_tokenizer = Whitespace()
+        trainer = BpeTrainer(vocab_size=self.config["VOCAB_SIZE"], special_tokens=self.special_tokens)
+        tokenizer.train_from_iterator(self.get_text_iterator(), trainer=trainer)
+        tokenizer.save(str(self.tokenizer_path)); logging.info(f"Tokenizer trained and saved to {self.tokenizer_path}")
+        return tokenizer
+    def get_tokenizer(self):
+        if not self.tokenizer_path.exists(): return self.train_tokenizer()
+        logging.info(f"Loading existing tokenizer from {self.tokenizer_path}")
+        return Tokenizer.from_file(str(self.tokenizer_path))
+
+# --- Dataset and Dataloader ---
+class OpusDataset(Dataset):
+    # ... (No changes needed in this class)
+    def __init__(self, tokenizer, config, split="train"):
+        self.tokenizer = tokenizer; self.config = config
+        dataset = load_dataset(f"Helsinki-NLP/opus-100", f"{config['SRC_LANG']}-{config['TGT_LANG']}", split=split)
+        self.pairs = [item['translation'] for item in dataset]
+        self.src_lang, self.tgt_lang, self.max_len = config["SRC_LANG"], config["TGT_LANG"], config["MAX_SEQ_LEN"]
+        self.bos_id, self.eos_id, self.pad_id = tokenizer.token_to_id("<s>"), tokenizer.token_to_id("</s>"), tokenizer.token_to_id("<pad>")
+    def __len__(self): return len(self.pairs)
+    def __getitem__(self, idx):
+        pair = self.pairs[idx]
+        src_text, tgt_text = pair[self.src_lang], pair[self.tgt_lang]
+        src_tokens = [self.bos_id] + self.tokenizer.encode(src_text).ids + [self.eos_id]
+        tgt_tokens = [self.bos_id] + self.tokenizer.encode(tgt_text).ids + [self.eos_id]
+        return {"src": torch.tensor(src_tokens[:self.max_len], dtype=torch.long), "tgt": torch.tensor(tgt_tokens[:self.max_len], dtype=torch.long)}
+
+class PadCollate:
+    # ... (No changes needed in this class)
+    def __init__(self, pad_id): self.pad_id = pad_id
+    def __call__(self, batch):
+        src_batch, tgt_batch = [item["src"] for item in batch], [item["tgt"] for item in batch]
+        src_padded = pad_sequence(src_batch, batch_first=True, padding_value=self.pad_id)
+        tgt_padded = pad_sequence(tgt_batch, batch_first=True, padding_value=self.pad_id)
+        return {"src": src_padded, "tgt": tgt_padded}
+
+# --- Model Architecture ---
+class PositionalEncoding(nn.Module):
+
+    def __init__(self, dim, dropout, max_len=5000):
+        super().__init__(); self.dropout = nn.Dropout(p=dropout)
+        position = torch.arange(max_len).unsqueeze(1); div_term = torch.exp(torch.arange(0, dim, 2) * (-math.log(10000.0) / dim))
+        pe = torch.zeros(max_len, 1, dim); pe[:, 0, 0::2] = torch.sin(position * div_term); pe[:, 0, 1::2] = torch.cos(position * div_term)
+        self.register_buffer('pe', pe)
+    def forward(self, x): x = x + self.pe[:x.size(0)]; return self.dropout(x)
+
+class TranslationTransformer(nn.Module):
+
+    def __init__(self, vocab_size, dim, n_heads, encoder_layers, decoder_layers, ff_dim, dropout, max_len):
+        super().__init__()
+        self.embedding = nn.Embedding(vocab_size, dim); self.pos_encoder = PositionalEncoding(dim, dropout, max_len)
+        self.transformer = nn.Transformer(d_model=dim, nhead=n_heads, num_encoder_layers=encoder_layers, num_decoder_layers=decoder_layers, dim_feedforward=ff_dim, dropout=dropout, batch_first=True)
+        self.generator = nn.Linear(dim, vocab_size)
+    def _generate_mask(self, src, tgt, pad_id):
+        tgt_mask = self.transformer.generate_square_subsequent_mask(tgt.shape[1], device=tgt.device)
+        src_padding_mask, tgt_padding_mask = (src == pad_id), (tgt == pad_id)
+        return tgt_mask, src_padding_mask, tgt_padding_mask
+    def forward(self, src, tgt, pad_id):
+        src_emb = self.pos_encoder((self.embedding(src) * math.sqrt(CONFIG["DIM"])).permute(1, 0, 2)).permute(1, 0, 2)
+        tgt_emb = self.pos_encoder((self.embedding(tgt) * math.sqrt(CONFIG["DIM"])).permute(1, 0, 2)).permute(1, 0, 2)
+        tgt_mask, src_padding_mask, tgt_padding_mask = self._generate_mask(src, tgt, pad_id)
+        output = self.transformer(src_emb, tgt_emb, tgt_mask=tgt_mask, src_key_padding_mask=src_padding_mask, tgt_key_padding_mask=tgt_padding_mask, memory_key_padding_mask=src_padding_mask)
+        return self.generator(output)
+
+# --- Trainer ---
+class Trainer:
+    def __init__(self, model, tokenizer, config):
+        self.model = model
+        self.tokenizer = tokenizer
+        self.config = config
+        self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+        self.model.to(self.device)
+        self.optimizer = torch.optim.AdamW(model.parameters(), lr=config["LEARNING_RATE"])
+        self.pad_id = tokenizer.token_to_id("<pad>")
+        self.criterion = nn.CrossEntropyLoss(ignore_index=self.pad_id)
+        self.scaler = torch.cuda.amp.GradScaler(enabled=(self.device.type == 'cuda'))
+        self.checkpoint_dir = Path(config["CHECKPOINT_DIR"])
+        self.checkpoint_dir.mkdir(exist_ok=True)
+
+    def train_epoch(self, dataloader):
+        self.model.train()
+        total_loss = 0
+        progress_bar = tqdm(dataloader, desc=f"Epoch {self.current_epoch+1}/{self.config['NUM_EPOCHS']} Training")
+        for batch in progress_bar:
+            src, tgt = batch["src"].to(self.device), batch["tgt"].to(self.device)
+            tgt_input, tgt_output = tgt[:, :-1], tgt[:, 1:]
+            self.optimizer.zero_grad(set_to_none=True)
+            with torch.amp.autocast(device_type=self.device.type, enabled=(self.device.type == 'cuda')):
+                logits = self.model(src, tgt_input, self.pad_id)
+                loss = self.criterion(logits.view(-1, logits.size(-1)), tgt_output.reshape(-1))
+            self.scaler.scale(loss).backward()
+            self.scaler.step(self.optimizer)
+            self.scaler.update()
+            total_loss += loss.item()
+            progress_bar.set_postfix({"loss": f"{loss.item():.4f}"})
+        return total_loss / len(dataloader)
+
+    # <<< NEW METHOD: For validation and testing >>>
+    def evaluate(self, dataloader, description="Evaluating"):
+        self.model.eval()
+        total_loss = 0
+        with torch.no_grad():
+            progress_bar = tqdm(dataloader, desc=description)
+            for batch in progress_bar:
+                src, tgt = batch["src"].to(self.device), batch["tgt"].to(self.device)
+                tgt_input, tgt_output = tgt[:, :-1], tgt[:, 1:]
+                logits = self.model(src, tgt_input, self.pad_id)
+                loss = self.criterion(logits.view(-1, logits.size(-1)), tgt_output.reshape(-1))
+                total_loss += loss.item()
+                progress_bar.set_postfix({"loss": f"{loss.item():.4f}"})
+        return total_loss / len(dataloader)
+
+    def save_checkpoint(self, epoch, val_loss):
+        filename = f"checkpoint_epoch_{epoch+1}_valloss_{val_loss:.4f}.pt"
+        path = self.checkpoint_dir / filename
+        torch.save({'epoch': epoch, 'model_state_dict': self.model.state_dict(),
+                    'optimizer_state_dict': self.optimizer.state_dict(), 'loss': val_loss}, path)
+        logging.info(f"Checkpoint saved to {path}")
+
+
+    def train(self, train_loader, val_loader):
+        for epoch in range(self.config["NUM_EPOCHS"]):
+            self.current_epoch = epoch
+            logging.info(f"--- Starting Epoch {epoch + 1}/{self.config['NUM_EPOCHS']} ---")
+            train_loss = self.train_epoch(train_loader)
+            val_loss = self.evaluate(val_loader, description=f"Epoch {epoch+1}/{self.config['NUM_EPOCHS']} Validation")
+            logging.info(f"Epoch {epoch+1} -> Train Loss: {train_loss:.4f}, Validation Loss: {val_loss:.4f}")
+            self.save_checkpoint(epoch, val_loss)
+            self.translate("This is a test to see how the model is learning.")
+
+    def translate(self, src_sentence: str):
+        self.model.eval()
+        src_tokens = [self.tokenizer.token_to_id("<s>")] + self.tokenizer.encode(src_sentence).ids + [self.tokenizer.token_to_id("</s>")]
+        src = torch.tensor(src_tokens, dtype=torch.long).unsqueeze(0).to(self.device)
+        tgt_tokens = [self.tokenizer.token_to_id("<s>")]
+        with torch.no_grad():
+            for _ in range(self.config["MAX_SEQ_LEN"]):
+                tgt_input = torch.tensor(tgt_tokens, dtype=torch.long).unsqueeze(0).to(self.device)
+                logits = self.model(src, tgt_input, self.pad_id)
+                next_token_id = logits[:, -1, :].argmax(dim=-1).item()
+                tgt_tokens.append(next_token_id)
+                if next_token_id == self.tokenizer.token_to_id("</s>"): break
+        translated_text = self.tokenizer.decode(tgt_tokens, skip_special_tokens=True)
+        logging.info(f"Source:      '{src_sentence}'")
+        logging.info(f"Translated:  '{translated_text}'")
+
+def main():
+    # Implemented a cuda check to see if my drivers are turning schizo again or not.
+    print("-" * 50)
+    print("CUDA Health Check:")
+    if torch.cuda.is_available():
+        print(f"✅ CUDA is available.")
+        print(f"   PyTorch Version: {torch.__version__}")
+        print(f"   CUDA Version PyTorch was built with: {torch.version.cuda}")
+        print(f"   Number of GPUs: {torch.cuda.device_count()}")
+        print(f"   Current GPU Name: {torch.cuda.get_device_name(0)}")
+    else:
+        print(f"❌ CUDA is NOT available.")
+        print(f"   PyTorch will run on CPU, which will be very slow.")
+        print(f"   ACTION: Ensure you have installed PyTorch with CUDA support. See https://pytorch.org/get-started/locally/")
+    print("-" * 50)
+
+    tokenizer_manager = TokenizerManager(CONFIG)
+    tokenizer = tokenizer_manager.get_tokenizer()
+    CONFIG["VOCAB_SIZE"] = tokenizer.get_vocab_size()
+    
+    logging.info("Loading and preparing datasets...")
+    train_dataset = OpusDataset(tokenizer, CONFIG, split="train")
+    val_dataset = OpusDataset(tokenizer, CONFIG, split="validation")
+    test_dataset = OpusDataset(tokenizer, CONFIG, split="test")
+    logging.info(f"Dataset sizes -> Train: {len(train_dataset)}, Validation: {len(val_dataset)}, Test: {len(test_dataset)}")
+    
+    pad_id = tokenizer.token_to_id("<pad>")
+    collate_fn = PadCollate(pad_id)
+    num_workers = 0 if os.name == 'nt' else os.cpu_count() // 2
+    
+    train_loader = DataLoader(train_dataset, batch_size=CONFIG["BATCH_SIZE"], shuffle=True, collate_fn=collate_fn, num_workers=num_workers, pin_memory=torch.cuda.is_available())
+    val_loader = DataLoader(val_dataset, batch_size=CONFIG["BATCH_SIZE"], shuffle=False, collate_fn=collate_fn, num_workers=num_workers, pin_memory=torch.cuda.is_available())
+    test_loader = DataLoader(test_dataset, batch_size=CONFIG["BATCH_SIZE"], shuffle=False, collate_fn=collate_fn, num_workers=num_workers, pin_memory=torch.cuda.is_available())
+
+    model = TranslationTransformer(vocab_size=CONFIG["VOCAB_SIZE"], dim=CONFIG["DIM"], n_heads=CONFIG["N_HEADS"],
+                                   encoder_layers=CONFIG["ENCODER_LAYERS"], decoder_layers=CONFIG["DECODER_LAYERS"],
+                                   ff_dim=CONFIG["FF_DIM"], dropout=CONFIG["DROPOUT"], max_len=CONFIG["MAX_SEQ_LEN"])
+    
+    total_params = sum(p.numel() for p in model.parameters() if p.requires_grad)
+    logging.info(f"Model initialized. Total trainable parameters: {total_params:,}")
+
+    trainer = Trainer(model, tokenizer, CONFIG)
+    
+
+    trainer.train(train_loader, val_loader)
+    
+    # NEW TESTS, NOT AS SHITTY AS BEFORE
+    logging.info("\n--- Training Complete. Evaluating on Test Set... ---")
+    test_loss = trainer.evaluate(test_loader, description="Final Test Evaluation")
+    logging.info(f"Final Test Loss: {test_loss:.4f}")
+
+    logging.info("\n--- Final Translation Examples ---")
+    trainer.translate("The European Economic Area was created in 1994.")
+    trainer.translate("What is your name?")
+    trainer.translate("This technology is changing the world.")
+
+if __name__ == "__main__":
+    main()

translator_loader.py

@@ -0,0 +1,195 @@
+import torch
+import torch.nn as nn
+from pathlib import Path
+import math
+import logging
+import re
+
+# --- Setup ---
+# Configure logging to be minimal for inference
+logging.basicConfig(level=logging.INFO, format='%(message)s')
+
+# --- Configuration (Must match the training script) ---
+CONFIG = {
+    "SRC_LANG": "en",
+    "TGT_LANG": "zh",
+    "TOKENIZER_FILE": "opus_en_zh_tokenizer.json",
+    "MAX_SEQ_LEN": 128,
+    "DIM": 256,
+    "ENCODER_LAYERS": 4,
+    "DECODER_LAYERS": 4,
+    "N_HEADS": 8,
+    "FF_DIM": 512,
+    "DROPOUT": 0.1,
+    "CHECKPOINT_DIR": "checkpoints_translation",
+}
+
+
+class PositionalEncoding(nn.Module):
+    def __init__(self, dim, dropout, max_len=5000):
+        super().__init__()
+        self.dropout = nn.Dropout(p=dropout)
+        position = torch.arange(max_len).unsqueeze(1)
+        div_term = torch.exp(torch.arange(0, dim, 2) * (-math.log(10000.0) / dim))
+        pe = torch.zeros(max_len, 1, dim)
+        pe[:, 0, 0::2] = torch.sin(position * div_term)
+        pe[:, 0, 1::2] = torch.cos(position * div_term)
+        self.register_buffer('pe', pe)
+
+    def forward(self, x):
+        x = x + self.pe[:x.size(0)]
+        return self.dropout(x)
+
+class TranslationTransformer(nn.Module):
+    def __init__(self, vocab_size, dim, n_heads, encoder_layers, decoder_layers, ff_dim, dropout, max_len):
+        super().__init__()
+        self.embedding = nn.Embedding(vocab_size, dim)
+        self.pos_encoder = PositionalEncoding(dim, dropout, max_len)
+        self.transformer = nn.Transformer(
+            d_model=dim, nhead=n_heads, num_encoder_layers=encoder_layers,
+            num_decoder_layers=decoder_layers, dim_feedforward=ff_dim,
+            dropout=dropout, batch_first=True
+        )
+        self.generator = nn.Linear(dim, vocab_size)
+
+    def _generate_mask(self, src, tgt, pad_id):
+        tgt_mask = self.transformer.generate_square_subsequent_mask(tgt.shape[1], device=tgt.device)
+        src_padding_mask = (src == pad_id)
+        tgt_padding_mask = (tgt == pad_id)
+        return tgt_mask, src_padding_mask, tgt_padding_mask
+
+    def forward(self, src, tgt, pad_id):
+        src_emb = self.pos_encoder((self.embedding(src) * math.sqrt(CONFIG["DIM"])).permute(1, 0, 2)).permute(1, 0, 2)
+        tgt_emb = self.pos_encoder((self.embedding(tgt) * math.sqrt(CONFIG["DIM"])).permute(1, 0, 2)).permute(1, 0, 2)
+        tgt_mask, src_padding_mask, tgt_padding_mask = self._generate_mask(src, tgt, pad_id)
+        output = self.transformer(
+            src_emb, tgt_emb, tgt_mask=tgt_mask, src_key_padding_mask=src_padding_mask,
+            tgt_key_padding_mask=tgt_padding_mask, memory_key_padding_mask=src_padding_mask
+        )
+        return self.generator(output)
+
+# We need to import the Tokenizer class to load the tokenizer file
+from tokenizers import Tokenizer
+
+class Translator:
+    def __init__(self, config):
+        self.config = config
+        self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+        logging.info(f"Using device: {self.device}")
+
+        # Load the trained tokenizer
+        tokenizer_path = Path(self.config["TOKENIZER_FILE"])
+        if not tokenizer_path.exists():
+            raise FileNotFoundError(f"Tokenizer file not found at {tokenizer_path}. Please run the training script first.")
+        self.tokenizer = Tokenizer.from_file(str(tokenizer_path))
+        
+        # Get special token IDs
+        self.bos_id = self.tokenizer.token_to_id("<s>")
+        self.eos_id = self.tokenizer.token_to_id("</s>")
+        self.pad_id = self.tokenizer.token_to_id("<pad>")
+
+        # Initialize the model structure
+        self.model = TranslationTransformer(
+            vocab_size=self.tokenizer.get_vocab_size(),
+            dim=self.config["DIM"], n_heads=self.config["N_HEADS"],
+            encoder_layers=self.config["ENCODER_LAYERS"], decoder_layers=self.config["DECODER_LAYERS"],
+            ff_dim=self.config["FF_DIM"], dropout=self.config["DROPOUT"], max_len=self.config["MAX_SEQ_LEN"]
+        )
+        self.model.to(self.device)
+
+    def load_best_checkpoint(self):
+        """Finds and loads the checkpoint with the lowest validation loss."""
+        checkpoint_dir = Path(self.config["CHECKPOINT_DIR"])
+        if not checkpoint_dir.exists():
+            raise FileNotFoundError(f"Checkpoint directory not found at {checkpoint_dir}.")
+
+        best_loss = float('inf')
+        best_checkpoint_path = None
+        
+        for chk_path in checkpoint_dir.glob("*.pt"):
+            # Use regex to find the validation loss in the filename
+            match = re.search(r'valloss_([\d.]+)\.pt', chk_path.name)
+            if match:
+                val_loss = float(match.group(1))
+                if val_loss < best_loss:
+                    best_loss = val_loss
+                    best_checkpoint_path = chk_path
+        
+        if best_checkpoint_path is None:
+            raise FileNotFoundError(f"No valid checkpoints found in {checkpoint_dir}. Checkpoint names must be like '...valloss_x.xxxx.pt'.")
+
+        logging.info(f"Loading best model from: {best_checkpoint_path} (Validation Loss: {best_loss:.4f})")
+        checkpoint = torch.load(best_checkpoint_path, map_location=self.device)
+        self.model.load_state_dict(checkpoint['model_state_dict'])
+        
+        # Set the model to evaluation mode. This is crucial!
+        # It disables layers like Dropout for consistent inference.
+        self.model.eval()
+
+    def translate(self, src_sentence: str):
+        """Translates a single English sentence to Chinese using greedy decoding."""
+        if not src_sentence.strip():
+            return ""
+
+        # Prepare the input
+        src_tokens = [self.bos_id] + self.tokenizer.encode(src_sentence).ids + [self.eos_id]
+        src = torch.tensor(src_tokens, dtype=torch.long).unsqueeze(0).to(self.device)
+        
+        # Start decoding
+        tgt_tokens = [self.bos_id]
+        
+        with torch.no_grad(): # Disable gradient calculation for efficiency
+            for _ in range(self.config["MAX_SEQ_LEN"]):
+                tgt_input = torch.tensor(tgt_tokens, dtype=torch.long).unsqueeze(0).to(self.device)
+                
+                # Get model predictions
+                logits = self.model(src, tgt_input, self.pad_id)
+                
+                # Get the most likely next token (greedy decoding)
+                next_token_id = logits[:, -1, :].argmax(dim=-1).item()
+                tgt_tokens.append(next_token_id)
+                
+                # Stop if the end-of-sentence token is generated
+                if next_token_id == self.eos_id:
+                    break
+        
+        # Decode the generated token IDs back to a string
+        translated_text = self.tokenizer.decode(tgt_tokens, skip_special_tokens=True)
+        return translated_text
+
+def interactive_session():
+    """Runs the main interactive translation loop."""
+    try:
+        translator = Translator(CONFIG)
+        translator.load_best_checkpoint()
+    except FileNotFoundError as e:
+        logging.error(f"Error initializing translator: {e}")
+        logging.error("Please make sure you have run the training script and have a valid tokenizer and checkpoint file.")
+        return
+
+    print("\n--- ZHEN - 1 Translator ---")
+    print("Type an English sentence and press Enter.")
+    print("Type 'quit' or 'exit' to close the program.")
+    
+    while True:
+        try:
+            source_text = input("\nEnglish > ")
+            if source_text.lower() in ['quit', 'exit', 'q']:
+                print("Exiting translator. Goodbye!")
+                break
+            
+            if not source_text:
+                continue
+
+            translated_text = translator.translate(source_text)
+            print(f"Chinese < {translated_text}")
+
+        except KeyboardInterrupt:
+            print("\nExiting translator. Goodbye!")
+            break
+        except Exception as e:
+            logging.error(f"An unexpected error occurred: {e}")
+
+
+if __name__ == "__main__":
+    interactive_session()

verify_cuda.py

@@ -0,0 +1,13 @@
+import torch
+
+print(f"PyTorch version: {torch.__version__}")
+print(f"CUDA available: {torch.cuda.is_available()}")
+
+if torch.cuda.is_available():
+    print(f"CUDA version PyTorch was built with: {torch.version.cuda}")
+    print(f"Number of GPUs: {torch.cuda.device_count()}")
+    print(f"Current device: {torch.cuda.current_device()}")
+    print(f"Device name: {torch.cuda.get_device_name(torch.cuda.current_device())}")
+else:
+    print("\n❌ PyTorch cannot find CUDA.")
+    print("   Follow the 'Foolproof Plan' to fix your environment.")