fix: encoding binary now SUPAH fast

src/process.py

@@ -62,7 +62,7 @@ def compress(
     print("Getting encoded value")
     interval_min, interval_max, _ = AE.get_encoded_value(stage)
     print("Encoding in binary")
-    binary_code, _ = AE.encode_binary(interval_min, interval_max)
+    binary_code, _ = AE.custom_binary_encoding(interval_min, interval_max)
 
     # Pack
     bits = binary_code.split(".", maxsplit=1)[1]
@@ -77,5 +77,31 @@ def compress(
         print(out_bytes)
 
 
-def decompress():
+def decompress(
-    return NotImplementedError("Decompression is not implemented yet")
+        device,
+        model_path: str,
+        input_file: str,
+        output_file: str | None = None
+):
+    context_length = 128
+    output = bytearray()
+
+    print("Reading in the data")
+    with open(input_file, "rb") as f:
+        bytes_data = f.read()
+
+    if len(bytes_data) == 0:
+        print("Input file is empty, nothing has to be done...")
+        return
+
+    print("Loading the model")
+    model = torch.load(model_path, weights_only=False)
+    model.to(device)
+    model.eval()
+
+    print("Decompressing")
+    ae = ArithmeticEncoding(frequency_table={0: 1})
+    stage_min, stage_max = Decimal(0), Decimal(1)
+
+    context = deque([0] * context_length, maxlen=context_length)
+

src/utils/custom_ae.py

@@ -201,40 +201,23 @@ class CustomArithmeticEncoding:
         float_interval_max: float
         """
         code = []
-        k = 1
+        found = False
-        halves = [
+        next_n = 0.5
-            [0.0, 1 / 2],
+        n = 0
-            [1 / 2, 1.0]
-        ]
 
-        i = 0
+        while not found:
-
+            if n + next_n < float_interval_max:
-        while i < 1024:
-            k += 1
-            i += 1
-
-            if halves[0][0] >= float_interval_min and halves[0][1] < float_interval_max:
-                break
-            if halves[1][0] >= float_interval_min and halves[1][1] < float_interval_max:
-                break
-
-            # left interval, insert 0
-            if float_interval_max < halves[0][1]:
-                code.append(0)
-                low = halves[0][0]
-                high = halves[0][1]
-
-            else:
                 code.append(1)
-                low = halves[1][0]
+                n += next_n
-                high = halves[1][1]
 
-            halves[0][0] = low
+                if n >= float_interval_min:
-            halves[0][1] = low + 1 / (1 << k)
+                    found = True
-            halves[1][0] = halves[0][1]
+            else:
-            halves[1][1] = high
+                code.append(0)
 
-        return "0." + ''.join(map(str, code)), k
+            next_n /= 2
+
+        return ''.join(map(str, code))
 
     def decode(self, encoded_msg, msg_length, probability_table):
         """
@@ -358,8 +341,8 @@ def bin2float(bin_num):
 if __name__ == "__main__":
     coder = CustomArithmeticEncoding({})
 
-    low = 0.25
+    low = 0.00324
-    high = 0.5
+    high = 0.357
 
     # slow_code = coder.encode_binary(low, high)
     fast_code = coder.custom_binary_encoding(low, high)